Posts List Oldapi View
We shared this request example with FAB participants: url_qparams = { "limit": count, "offset": offset, "has_group": "false", "order_by": "-activity", "forecast_type": "binary", "project": tournament_id, "status": "open", "type": "forecast", "include_description": "true", } url = f"{api_info.base_url}/questions/" response = requests.get( url, headers={"Authorization": f"Token {api_info.token}"}, params=url_qparams )
But we don't want to support all these parameters, and the ones relevant are: - order_by - status - project - forecast_type - we ignore this, but assume it's binary - FAB only supports binary for now.
GET /api2/questions/?format=api&offset=2920
{ "count": 5974, "next": "http://www.metaculus.com/api2/questions/?format=api&limit=20&offset=2940", "previous": "http://www.metaculus.com/api2/questions/?format=api&limit=20&offset=2900", "results": [ { "id": 20278, "title": "Will any 100% AI-generated legislation be passed in the U.S. before 2040?", "short_title": "AI-generated US legislation by 2040", "url_title": "AI-generated US legislation by 2040", "slug": "ai-generated-us-legislation-by-2040", "author_id": 111743, "author_username": "NathanpmYoung", "coauthors": [], "created_at": "2023-12-01T18:28:42.793979Z", "published_at": "2023-12-08T12:48:00Z", "edited_at": "2025-09-12T15:42:56.277495Z", "curation_status": "approved", "curation_status_updated_at": "2023-12-08T12:48:00Z", "comment_count": 11, "status": "open", "resolved": false, "actual_close_time": null, "scheduled_close_time": "2039-12-31T00:00:00Z", "scheduled_resolve_time": "2040-01-08T00:00:00Z", "actual_resolve_time": null, "open_time": "2023-12-08T12:48:00Z", "nr_forecasters": 35, "html_metadata_json": null, "projects": { "topic": [ { "id": 15869, "name": "Artificial Intelligence", "slug": "ai", "emoji": "🤖", "type": "topic" } ], "site_main": [ { "id": 144, "type": "site_main", "name": "Metaculus Community", "slug": null, "header_image": null, "prize_pool": null, "start_date": null, "close_date": null, "forecasting_end_date": null, "html_metadata_json": null, "is_ongoing": null, "user_permission": "forecaster", "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2025-07-18T17:28:18.838588Z", "score_type": null, "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } ], "default_project": { "id": 144, "type": "site_main", "name": "Metaculus Community", "slug": null, "header_image": null, "prize_pool": null, "start_date": null, "close_date": null, "forecasting_end_date": null, "html_metadata_json": null, "is_ongoing": null, "user_permission": "forecaster", "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2025-07-18T17:28:18.838588Z", "score_type": null, "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" }, "category": [ { "id": 3692, "name": "Computing and Math", "slug": "computing-and-math", "emoji": "💻", "description": "Computing and Math", "type": "category" }, { "id": 3694, "name": "Artificial Intelligence", "slug": "artificial-intelligence", "emoji": "🤖", "description": "Artificial Intelligence", "type": "category" } ] }, "question": { "id": 20278, "title": "Will any 100% AI-generated legislation be passed in the U.S. before 2040?", "created_at": "2023-12-01T18:28:42.793979Z", "open_time": "2023-12-08T12:48:00Z", "cp_reveal_time": "2023-12-11T12:48:00Z", "spot_scoring_time": "2023-12-11T12:48:00Z", "scheduled_resolve_time": "2040-01-08T00:00:00Z", "actual_resolve_time": null, "resolution_set_time": null, "scheduled_close_time": "2039-12-31T00:00:00Z", "actual_close_time": "2039-12-31T00:00:00Z", "type": "binary", "options": null, "group_variable": "", "status": "open", "possibilities": { "type": "binary" }, "resolution": null, "include_bots_in_aggregates": false, "question_weight": 1.0, "default_score_type": "peer", "default_aggregation_method": "recency_weighted", "label": "", "unit": "", "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "scaling": { "range_min": null, "range_max": null, "nominal_min": null, "nominal_max": null, "zero_point": null, "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "continuous_range": null }, "group_rank": null, "description": "AI has been rapidly developing in recent years, and well before this it has long been opined that decisions in our legal and political systems could be better handled by AI than human beings. The passage of legislation, authored entirely by AI, would represent a major milestone toward such a future.", "resolution_criteria": "The question resolves as **Yes** if, before the start of the year 2040, it is publicly reported by credible sources that an AI has entirely written binding legislation that has been passed by a legislative body at the local, state, and/or federal level in the United States. \n\n\"Binding legislation\" shall include official budgets, statutes/ordinances, and amendments to constitutions or charters that are voted on by an elected legislative body. Official statements or resolutions that impose no material change to any persons, legal or natural, shall not be counted. \n\nLegislation is considered \"entirely written\" by the AI system if every word of the legislation is generated by an AI, including in response to human prompts.\n\nIf this does not happen before January 1, 2040, the question resolves **No**.", "fine_print": "", "post_id": 20278, "aggregations": { "recency_weighted": { "history": [ { "start_time": 1757639964.918954, "end_time": 1762056074.913278, "forecaster_count": 28, "interval_lower_bounds": [ 0.16 ], "centers": [ 0.33 ], "interval_upper_bounds": [ 0.55 ] } ], "latest": { "start_time": 1757639964.918954, "end_time": 1762056074.913278, "forecaster_count": 28, "interval_lower_bounds": [ 0.16 ], "centers": [ 0.33 ], "interval_upper_bounds": [ 0.55 ], "forecast_values": [ 0.6699999999999999, 0.33 ], "means": [ 0.38281937608550204 ], "histogram": [ [ 0.0, 0.31302422044387057, 0.0, 0.0, 0.0, 0.2720306930742662, 0.0, 0.0, 0.0, 0.0, 0.10111441089353605, 0.9462524413216661, 0.0, 0.0, 0.0, 0.3108648595307786, 0.44071068269949837, 0.0, 0.013684347437579939, 0.0, 1.0, 0.0, 0.0, 0.0, 0.160831023763914, 0.0, 0.35033622771640044, 0.0, 0.5482154774436659, 0.0, 0.0, 0.0, 0.0, 0.8859183200045558, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.9100808516300489, 0.3935277555318061, 0.0, 0.0, 0.0, 0.0, 0.6753507208774424, 0.0, 0.0, 0.0, 0.0, 0.33121025782770785, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.2748574657272404, 0.0, 0.0, 0.0, 0.0, 0.020706884286877357, 0.0, 0.0, 0.0, 0.0, 0.04710224358329484, 0.0, 0.0, 0.0, 0.0, 0.4921720506514511, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.6091619622184999, 0.0, 0.0, 0.0, 0.0 ] ] }, "score_data": {}, "movement": null }, "unweighted": { "history": [], "latest": null, "score_data": {}, "movement": null }, "single_aggregation": { "history": [], "latest": null, "score_data": {}, "movement": null }, "metaculus_prediction": { "history": [ { "start_time": 1728288418.765203, "end_time": null, "forecaster_count": 31, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null } ], "latest": { "start_time": 1728288418.765203, "end_time": null, "forecaster_count": 31, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null, "forecast_values": [ 0.9135364501564572, 0.0864635498435428 ], "means": null, "histogram": null }, "score_data": {}, "movement": null } } }, "user_permission": "forecaster", "vote": { "score": 0, "user_vote": null }, "forecasts_count": 109, "key_factors": [], "is_current_content_translated": false, "description": "AI has been rapidly developing in recent years, and well before this it has long been opined that decisions in our legal and political systems could be better handled by AI than human beings. The passage of legislation, authored entirely by AI, would represent a major milestone toward such a future." }, { "id": 20261, "title": "Will any additional Republican candidates for president drop out before 2024?", "short_title": "GOP Presidential Drop Out Before 2024?", "url_title": "GOP Presidential Drop Out Before 2024?", "slug": "gop-presidential-drop-out-before-2024", "author_id": 117502, "author_username": "RyanBeck", "coauthors": [], "created_at": "2023-11-29T21:53:33.570303Z", "published_at": "2023-11-30T15:30:00Z", "edited_at": "2025-09-05T17:28:53.247992Z", "curation_status": "approved", "curation_status_updated_at": "2023-11-30T15:30:00Z", "comment_count": 5, "status": "resolved", "resolved": true, "actual_close_time": "2023-12-04T16:04:00Z", "scheduled_close_time": "2023-12-31T23:00:00Z", "scheduled_resolve_time": "2023-12-31T23:00:00Z", "actual_resolve_time": "2023-12-04T16:04:00Z", "open_time": "2023-11-30T15:30:00Z", "nr_forecasters": 38, "html_metadata_json": null, "projects": { "leaderboard_tag": [ { "id": 32591, "name": "2023 Leaderboard", "slug": "2023_leaderboard", "type": "leaderboard_tag" } ], "site_main": [ { "id": 144, "type": "site_main", "name": "Metaculus Community", "slug": null, "header_image": null, "prize_pool": null, "start_date": null, "close_date": null, "forecasting_end_date": null, "html_metadata_json": null, "is_ongoing": null, "user_permission": "forecaster", "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2025-07-18T17:28:18.838588Z", "score_type": null, "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } ], "category": [ { "id": 3685, "name": "Elections", "slug": "elections", "emoji": "🗳️", "description": "Elections", "type": "category" }, { "id": 3689, "name": "Politics", "slug": "politics", "emoji": "🏛️", "description": "Politics", "type": "category" } ], "tournament": [ { "id": 2654, "type": "tournament", "name": "🏆 Q4 2023 Quarterly Cup 🏆", "slug": "quarterly-cup-2023q4", "header_image": "https://metaculus-web-media.s3.amazonaws.com/quarterlycup.webp", "prize_pool": null, "start_date": "2023-10-03T15:00:00Z", "close_date": "2024-01-04T12:00:00Z", "forecasting_end_date": "2023-12-31T23:00:00Z", "html_metadata_json": null, "is_ongoing": false, "user_permission": null, "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2024-01-05T15:54:55.645171Z", "score_type": "relative_legacy_tournament", "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } ], "default_project": { "id": 2654, "type": "tournament", "name": "🏆 Q4 2023 Quarterly Cup 🏆", "slug": "quarterly-cup-2023q4", "header_image": "https://metaculus-web-media.s3.amazonaws.com/quarterlycup.webp", "prize_pool": null, "start_date": "2023-10-03T15:00:00Z", "close_date": "2024-01-04T12:00:00Z", "forecasting_end_date": "2023-12-31T23:00:00Z", "html_metadata_json": null, "is_ongoing": false, "user_permission": null, "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2024-01-05T15:54:55.645171Z", "score_type": "relative_legacy_tournament", "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } }, "question": { "id": 20261, "title": "Will any additional Republican candidates for president drop out before 2024?", "created_at": "2023-11-29T21:53:33.570303Z", "open_time": "2023-11-30T15:30:00Z", "cp_reveal_time": "2023-12-03T15:30:00Z", "spot_scoring_time": "2023-12-03T15:30:00Z", "scheduled_resolve_time": "2023-12-31T23:00:00Z", "actual_resolve_time": "2023-12-04T16:04:00Z", "resolution_set_time": "2023-12-04T16:04:00Z", "scheduled_close_time": "2023-12-31T23:00:00Z", "actual_close_time": "2023-12-04T16:04:00Z", "type": "binary", "options": null, "group_variable": "", "status": "resolved", "possibilities": { "type": "binary" }, "resolution": "yes", "include_bots_in_aggregates": false, "question_weight": 1.0, "default_score_type": "peer", "default_aggregation_method": "recency_weighted", "label": "", "unit": "", "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "scaling": { "range_min": null, "range_max": null, "nominal_min": null, "nominal_max": null, "zero_point": null, "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "continuous_range": null }, "group_rank": null, "description": "The Republican primary election will kick off [on January 15, 2024](https://www.cnn.com/election/2024/calendar), with the Iowa caucus, followed soon after by the New Hampshire primary on January 23rd. Several candidates have already [dropped out](https://en.wikipedia.org/wiki/2024_Republican_Party_presidential_primaries#Withdrew_before_the_primaries), the most recent being Tim Scott on November 12th. [FiveThirtyEight shows seven major candidates remaining](https://projects.fivethirtyeight.com/polls/president-primary-r/2024/national/) though Donald Trump is polling at roughly 60% as of November 29th and has so far his lead has been stable.", "resolution_criteria": "This question will resolve as **Yes** if, after November 29, 2023, and before January 1, 2024, any of the individuals below seeking the nomination as the Republican candidate for president announces that they are dropping out of the race:\n\n* Donald Trump\n* Ron DeSantis\n* Nikki Haley\n* Vivek Ramaswamy\n* Chris Christie\n* Doug Burgum\n* Asa Hutchinson", "fine_print": "", "post_id": 20261, "aggregations": { "recency_weighted": { "history": [ { "start_time": 1701706051.95233, "end_time": null, "forecaster_count": 38, "interval_lower_bounds": [ 0.35 ], "centers": [ 0.5 ], "interval_upper_bounds": [ 0.6 ] } ], "latest": { "start_time": 1701706051.95233, "end_time": null, "forecaster_count": 38, "interval_lower_bounds": [ 0.35 ], "centers": [ 0.5 ], "interval_upper_bounds": [ 0.6 ], "forecast_values": [ 0.5, 0.5 ], "means": [ 0.49462782206536343 ], "histogram": [ [ 0.0, 0.0, 0.0, 0.04223888642530301, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.035579456289155105, 0.0, 0.0, 0.0, 1.98762653538766, 0.0, 0.0, 0.0, 0.0, 0.14634721205825776, 0.0, 0.0, 0.0, 0.0, 0.25446741675480405, 0.0, 0.0, 0.06718452183420129, 0.0, 0.4706366803324597, 0.0, 0.0, 0.0, 0.0, 0.5593033974205628, 0.0, 0.0, 0.0, 0.0, 0.848390712889754, 0.0, 0.42198459712074377, 0.0, 0.0, 2.3183201556293342, 0.0, 0.16438976427023788, 0.0, 0.0, 0.37974261668515136, 0.0, 0.0, 0.0, 0.0, 1.248809595753717, 0.0, 0.0, 0.0, 0.0, 0.18409965808495726, 0.019676189570906492, 0.029639045211680565, 0.0, 0.0, 0.1011216878911307, 0.005716411657967498, 0.0, 0.0, 0.0, 0.20559680035186842, 0.0, 0.28211668489493613, 0.0, 0.0, 0.06521963758888388, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 1.0 ] ] }, "score_data": { "peer_score": 3.345484675935418, "coverage": 0.12830444421120984, "baseline_score": 0.7760396093739695, "spot_peer_score": 29.810733470873757, "peer_archived_score": 3.345484675935418, "baseline_archived_score": 0.7760396093739695, "spot_peer_archived_score": 29.810733470873757 }, "movement": null }, "unweighted": { "history": [], "latest": null, "score_data": {}, "movement": null }, "single_aggregation": { "history": [], "latest": null, "score_data": {}, "movement": null }, "metaculus_prediction": { "history": [ { "start_time": 1701684000.27332, "end_time": null, "forecaster_count": 38, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null } ], "latest": { "start_time": 1701684000.27332, "end_time": null, "forecaster_count": 38, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null, "forecast_values": [ 0.6788990352725978, 0.32110096472740224 ], "means": null, "histogram": null }, "score_data": {}, "movement": null } } }, "user_permission": "forecaster", "vote": { "score": 6, "user_vote": null }, "forecasts_count": 74, "key_factors": [], "is_current_content_translated": false, "description": "The Republican primary election will kick off [on January 15, 2024](https://www.cnn.com/election/2024/calendar), with the Iowa caucus, followed soon after by the New Hampshire primary on January 23rd. Several candidates have already [dropped out](https://en.wikipedia.org/wiki/2024_Republican_Party_presidential_primaries#Withdrew_before_the_primaries), the most recent being Tim Scott on November 12th. [FiveThirtyEight shows seven major candidates remaining](https://projects.fivethirtyeight.com/polls/president-primary-r/2024/national/) though Donald Trump is polling at roughly 60% as of November 29th and has so far his lead has been stable." }, { "id": 20260, "title": "Will a majority of voters approve Venezuela's referendum on incorporating Guayana Esequiba into Venezuela?", "short_title": "Venezuela Votes to Annex Guyana Territory?", "url_title": "Venezuela Votes to Annex Guyana Territory?", "slug": "venezuela-votes-to-annex-guyana-territory", "author_id": 117502, "author_username": "RyanBeck", "coauthors": [], "created_at": "2023-11-29T20:33:37.122683Z", "published_at": "2023-11-30T00:00:00Z", "edited_at": "2025-09-05T17:28:58.041041Z", "curation_status": "approved", "curation_status_updated_at": "2023-11-30T00:00:00Z", "comment_count": 14, "status": "resolved", "resolved": true, "actual_close_time": "2023-12-02T23:00:00Z", "scheduled_close_time": "2023-12-02T23:00:00Z", "scheduled_resolve_time": "2023-12-04T04:38:00Z", "actual_resolve_time": "2023-12-04T04:38:00Z", "open_time": "2023-11-30T00:00:00Z", "nr_forecasters": 62, "html_metadata_json": null, "projects": { "leaderboard_tag": [ { "id": 32591, "name": "2023 Leaderboard", "slug": "2023_leaderboard", "type": "leaderboard_tag" } ], "site_main": [ { "id": 144, "type": "site_main", "name": "Metaculus Community", "slug": null, "header_image": null, "prize_pool": null, "start_date": null, "close_date": null, "forecasting_end_date": null, "html_metadata_json": null, "is_ongoing": null, "user_permission": "forecaster", "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2025-07-18T17:28:18.838588Z", "score_type": null, "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } ], "category": [ { "id": 3685, "name": "Elections", "slug": "elections", "emoji": "🗳️", "description": "Elections", "type": "category" }, { "id": 3687, "name": "Geopolitics", "slug": "geopolitics", "emoji": "🌍", "description": "Geopolitics", "type": "category" } ], "tournament": [ { "id": 2654, "type": "tournament", "name": "🏆 Q4 2023 Quarterly Cup 🏆", "slug": "quarterly-cup-2023q4", "header_image": "https://metaculus-web-media.s3.amazonaws.com/quarterlycup.webp", "prize_pool": null, "start_date": "2023-10-03T15:00:00Z", "close_date": "2024-01-04T12:00:00Z", "forecasting_end_date": "2023-12-31T23:00:00Z", "html_metadata_json": null, "is_ongoing": false, "user_permission": null, "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2024-01-05T15:54:55.645171Z", "score_type": "relative_legacy_tournament", "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } ], "default_project": { "id": 2654, "type": "tournament", "name": "🏆 Q4 2023 Quarterly Cup 🏆", "slug": "quarterly-cup-2023q4", "header_image": "https://metaculus-web-media.s3.amazonaws.com/quarterlycup.webp", "prize_pool": null, "start_date": "2023-10-03T15:00:00Z", "close_date": "2024-01-04T12:00:00Z", "forecasting_end_date": "2023-12-31T23:00:00Z", "html_metadata_json": null, "is_ongoing": false, "user_permission": null, "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2024-01-05T15:54:55.645171Z", "score_type": "relative_legacy_tournament", "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } }, "question": { "id": 20260, "title": "Will a majority of voters approve Venezuela's referendum on incorporating Guayana Esequiba into Venezuela?", "created_at": "2023-11-29T20:33:37.122683Z", "open_time": "2023-11-30T00:00:00Z", "cp_reveal_time": "2023-11-30T07:06:00Z", "spot_scoring_time": "2023-11-30T07:06:00Z", "scheduled_resolve_time": "2023-12-04T04:38:00Z", "actual_resolve_time": "2023-12-04T04:38:00Z", "resolution_set_time": "2023-12-04T04:38:00Z", "scheduled_close_time": "2023-12-02T23:00:00Z", "actual_close_time": "2023-12-02T23:00:00Z", "type": "binary", "options": null, "group_variable": "", "status": "resolved", "possibilities": { "type": "binary" }, "resolution": "yes", "include_bots_in_aggregates": false, "question_weight": 1.0, "default_score_type": "peer", "default_aggregation_method": "recency_weighted", "label": "", "unit": "", "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "scaling": { "range_min": null, "range_max": null, "nominal_min": null, "nominal_max": null, "zero_point": null, "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "continuous_range": null }, "group_rank": null, "description": "A [Venezuelan referendum asking five questions is scheduled for December 3, 2023](https://en.wikipedia.org/wiki/2023_Venezuelan_referendum). The subject of the referendum is the territory of [Guayana Esequiba](https://en.wikipedia.org/wiki/Guayana_Esequiba), administered by Guyana but claimed by Venezuela. Guyana [has asked](https://www.icj-cij.org/case/171) the UN [International Court of Justice](https://en.wikipedia.org/wiki/International_Court_of_Justice) to halt the referendum, but Venezuela's Vice President [has stated](https://abcnews.go.com/International/wireStory/uns-top-court-venezuela-vows-press-ahead-referendum-104903839) \"Nothing will prevent the referendum scheduled for Dec. 3 from being held\".\n\n[According to Demerara Waves](https://demerarawaves.com/2023/10/23/venezuelan-electoral-council-approves-five-questions-for-referendum-on-essequibo/) the five questions asked in the referendum are as follows:\n\n1. Do you agree to reject by all means in accordance with the law the line fraudulently imposed by the Paris Arbitration Award of 1899 that seeks to deprive us of our Guayana Esequiba?\n\n2. Do you support the 1966 Geneva Agreement as the only valid legal instrument to reach a practical and satisfactory solution for Venezuela and Guyana regarding the controversy over the territory of Guayana Esequiba?\n\n3. Do you agree with Venezuela’s historical position of not recognizing the jurisdiction of the International Court of Justice to resolve the territorial controversy over Guayana Esequiba?\n\n4. Do you agree to oppose by all means in accordance with the law Guyana’s claim to unilaterally dispose of a sea pending delimitation illegally and in violation of international law?\n\n5. Do you agree with the creation of the Guayana Esequiba state and the development of an accelerated plan for the comprehensive care of the current and future population of that territory that includes, among others, the granting of citizenship and Venezuelan identity card in accordance with the Geneva Agreement and international law, consequently incorporating said state on the map of Venezuelan territory?", "resolution_criteria": "This question will resolve as **Yes** if the fifth question in the December 3rd, 2023, Venezuelan referendum — regarding incorporating Guayana Esequiba into Venezuelan territory — receives more votes in favor than opposed, according to the results published by Venezuela and as reported by [credible sources](https://www.metaculus.com/help/faq/#definitions).", "fine_print": "* This question asks specifically about the results of the [following referendum question](https://demerarawaves.com/2023/10/23/venezuelan-electoral-council-approves-five-questions-for-referendum-on-essequibo/):\n\n>Do you agree with the creation of the Guayana Esequiba state and the development of an accelerated plan for the comprehensive care of the current and future population of that territory that includes, among others, the granting of citizenship and Venezuelan identity card in accordance with the Geneva Agreement and international law, consequently incorporating said state on the map of Venezuelan territory?\n\n* This question asks about the results that Venezuela reports. Whether those results are broadly seen as credible is immaterial.\n\n* If no referendum is held on December 3rd this question will be **annulled**.", "post_id": 20260, "aggregations": { "recency_weighted": { "history": [ { "start_time": 1701555771.761095, "end_time": null, "forecaster_count": 62, "interval_lower_bounds": [ 0.9 ], "centers": [ 0.93 ], "interval_upper_bounds": [ 0.96 ] } ], "latest": { "start_time": 1701555771.761095, "end_time": null, "forecaster_count": 62, "interval_lower_bounds": [ 0.9 ], "centers": [ 0.93 ], "interval_upper_bounds": [ 0.96 ], "forecast_values": [ 0.06999999999999995, 0.93 ], "means": [ 0.904038896868924 ], "histogram": [ [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.004407189337132444, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.2144205484641543, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.48066753395180806, 0.003560195227230837, 0.0, 0.0, 0.0, 0.1809392623709519, 0.0, 0.0, 0.0, 0.0, 0.006437720576519247, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 1.4490016412966167, 0.24826285979550225, 0.0, 0.0, 0.0, 0.20402669149047276, 0.03565285440045971, 0.31167199523019246, 0.0, 0.0, 1.2783580677078952, 0.7231248136457471, 0.0, 2.342775599832415, 0.2123642000108881, 0.9386644706653295, 2.845563371920317, 0.13705347335991425, 0.11890323980894008, 2.5179256720676464 ] ] }, "score_data": { "peer_score": 8.263151143134436, "coverage": 0.9999262579263468, "baseline_score": 84.73064769918093, "spot_peer_score": 2.7866685534558853, "peer_archived_score": 8.263151143134436, "baseline_archived_score": 84.73064769918093, "spot_peer_archived_score": 2.7866685534558853 }, "movement": null }, "unweighted": { "history": [], "latest": null, "score_data": {}, "movement": null }, "single_aggregation": { "history": [], "latest": null, "score_data": {}, "movement": null }, "metaculus_prediction": { "history": [ { "start_time": 1701555771.808048, "end_time": null, "forecaster_count": 62, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null } ], "latest": { "start_time": 1701555771.808048, "end_time": null, "forecaster_count": 62, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null, "forecast_values": [ 0.0565067256298486, 0.9434932743701514 ], "means": null, "histogram": null }, "score_data": {}, "movement": null } } }, "user_permission": "forecaster", "vote": { "score": 6, "user_vote": null }, "forecasts_count": 136, "key_factors": [], "is_current_content_translated": false, "description": "A [Venezuelan referendum asking five questions is scheduled for December 3, 2023](https://en.wikipedia.org/wiki/2023_Venezuelan_referendum). The subject of the referendum is the territory of [Guayana Esequiba](https://en.wikipedia.org/wiki/Guayana_Esequiba), administered by Guyana but claimed by Venezuela. Guyana [has asked](https://www.icj-cij.org/case/171) the UN [International Court of Justice](https://en.wikipedia.org/wiki/International_Court_of_Justice) to halt the referendum, but Venezuela's Vice President [has stated](https://abcnews.go.com/International/wireStory/uns-top-court-venezuela-vows-press-ahead-referendum-104903839) \"Nothing will prevent the referendum scheduled for Dec. 3 from being held\".\n\n[According to Demerara Waves](https://demerarawaves.com/2023/10/23/venezuelan-electoral-council-approves-five-questions-for-referendum-on-essequibo/) the five questions asked in the referendum are as follows:\n\n1. Do you agree to reject by all means in accordance with the law the line fraudulently imposed by the Paris Arbitration Award of 1899 that seeks to deprive us of our Guayana Esequiba?\n\n2. Do you support the 1966 Geneva Agreement as the only valid legal instrument to reach a practical and satisfactory solution for Venezuela and Guyana regarding the controversy over the territory of Guayana Esequiba?\n\n3. Do you agree with Venezuela’s historical position of not recognizing the jurisdiction of the International Court of Justice to resolve the territorial controversy over Guayana Esequiba?\n\n4. Do you agree to oppose by all means in accordance with the law Guyana’s claim to unilaterally dispose of a sea pending delimitation illegally and in violation of international law?\n\n5. Do you agree with the creation of the Guayana Esequiba state and the development of an accelerated plan for the comprehensive care of the current and future population of that territory that includes, among others, the granting of citizenship and Venezuelan identity card in accordance with the Geneva Agreement and international law, consequently incorporating said state on the map of Venezuelan territory?" }, { "id": 20257, "title": "Will Mark Zuckerberg remain as CEO of Meta Platforms until October 21, 2031?", "short_title": "Zuckerberg CEO of Meta until 2031", "url_title": "Zuckerberg CEO of Meta until 2031", "slug": "zuckerberg-ceo-of-meta-until-2031", "author_id": 111743, "author_username": "NathanpmYoung", "coauthors": [], "created_at": "2023-11-29T18:30:56.157386Z", "published_at": "2024-04-10T18:00:00Z", "edited_at": "2025-09-05T17:29:28.069325Z", "curation_status": "approved", "curation_status_updated_at": "2024-04-10T18:00:00Z", "comment_count": 4, "status": "open", "resolved": false, "actual_close_time": null, "scheduled_close_time": "2031-10-21T00:00:00Z", "scheduled_resolve_time": "2031-10-28T00:00:00Z", "actual_resolve_time": null, "open_time": "2024-04-10T18:00:00Z", "nr_forecasters": 15, "html_metadata_json": null, "projects": { "site_main": [ { "id": 144, "type": "site_main", "name": "Metaculus Community", "slug": null, "header_image": null, "prize_pool": null, "start_date": null, "close_date": null, "forecasting_end_date": null, "html_metadata_json": null, "is_ongoing": null, "user_permission": "forecaster", "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2025-07-18T17:28:18.838588Z", "score_type": null, "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } ], "default_project": { "id": 144, "type": "site_main", "name": "Metaculus Community", "slug": null, "header_image": null, "prize_pool": null, "start_date": null, "close_date": null, "forecasting_end_date": null, "html_metadata_json": null, "is_ongoing": null, "user_permission": "forecaster", "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2025-07-18T17:28:18.838588Z", "score_type": null, "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" }, "category": [ { "id": 3698, "name": "Economy & Business", "slug": "economy-business", "emoji": "💼", "description": "Economy & Business", "type": "category" } ] }, "question": { "id": 20257, "title": "Will Mark Zuckerberg remain as CEO of Meta Platforms until October 21, 2031?", "created_at": "2023-11-29T18:30:56.157386Z", "open_time": "2024-04-10T18:00:00Z", "cp_reveal_time": "2024-04-13T18:00:00Z", "spot_scoring_time": "2024-04-13T18:00:00Z", "scheduled_resolve_time": "2031-10-28T00:00:00Z", "actual_resolve_time": null, "resolution_set_time": null, "scheduled_close_time": "2031-10-21T00:00:00Z", "actual_close_time": "2031-10-21T00:00:00Z", "type": "binary", "options": null, "group_variable": "", "status": "open", "possibilities": { "type": "binary" }, "resolution": null, "include_bots_in_aggregates": false, "question_weight": 1.0, "default_score_type": "peer", "default_aggregation_method": "recency_weighted", "label": "", "unit": "", "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "scaling": { "range_min": null, "range_max": null, "nominal_min": null, "nominal_max": null, "zero_point": null, "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "continuous_range": null }, "group_rank": null, "description": "On October 21, 2022, Facebook, the social media powerhouse, rebranded to Meta Platforms, reflecting CEO Mark Zuckerberg's shift towards a virtual reality-focused future, termed the Metaverse. Following the rebranding, Meta's stock value dropped due to poor earnings, raising concerns about the firm's future profitability and relevance under Zuckerberg's new direction. While share prices have rebounded, ongoing financial struggles, challenges in making the Metaverse profitable, or doubts about the company's adaptability in the ever-changing social media landscape could potentially lead to the end of Zuckerberg's [leadership](https://about.meta.com/media-gallery/executives/) as CEO.\n\nSee also: \n\n- Washington Post: [With Bezos out as Amazon CEO, Zuckerberg is the last man standing](https://www.washingtonpost.com/technology/2021/02/03/bezos-zuckerberg-ceo-transition/)\n\n- Harvard Business Review: [Should Mark Zuckerberg Resign?](https://hbr.org/2019/01/should-mark-zuckerberg-resign)", "resolution_criteria": "This market will resolve as **Yes** if Mark Zuckerberg continuously serves as Chief Executive Officer (CEO) of Meta Platforms Inc. until the market's closure. It will also resolve as **Yes** if Meta merges or is acquired, provided Zuckerberg remains CEO of the resultant entity. However, if [credible sources](https://www.metaculus.com/help/faq/#definitions) report Zuckerberg is no longer CEO of Meta (or its successor entity), or if he steps down or is removed as CEO before October 21, 2031, the market will immediately resolve to **No**.", "fine_print": "", "post_id": 20257, "aggregations": { "recency_weighted": { "history": [ { "start_time": 1741124025.997429, "end_time": null, "forecaster_count": 14, "interval_lower_bounds": [ 0.5 ], "centers": [ 0.639 ], "interval_upper_bounds": [ 0.71 ] } ], "latest": { "start_time": 1741124025.997429, "end_time": null, "forecaster_count": 14, "interval_lower_bounds": [ 0.5 ], "centers": [ 0.639 ], "interval_upper_bounds": [ 0.71 ], "forecast_values": [ 0.361, 0.639 ], "means": [ 0.6269197317316266 ], "histogram": [ [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0644634174904052, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.6537484749236214, 0.0, 0.0, 0.0, 0.0, 0.0, 0.7576333089753814, 0.5094663623840316, 0.0, 0.0, 0.0, 0.0, 0.560245764355437, 0.0, 0.0, 0.0, 0.0, 0.22188647333713657, 0.0, 0.0, 0.47632380816539127, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.4012260631462153, 0.0, 0.0, 0.0, 0.2315861692124475, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.8727500189273638, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.27467474026865935 ] ] }, "score_data": {}, "movement": null }, "unweighted": { "history": [], "latest": null, "score_data": {}, "movement": null }, "single_aggregation": { "history": [], "latest": null, "score_data": {}, "movement": null }, "metaculus_prediction": { "history": [ { "start_time": 1728290005.412696, "end_time": null, "forecaster_count": 14, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null } ], "latest": { "start_time": 1728290005.412696, "end_time": null, "forecaster_count": 14, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null, "forecast_values": [ 0.4509148535391727, 0.5490851464608273 ], "means": null, "histogram": null }, "score_data": {}, "movement": null } } }, "user_permission": "forecaster", "vote": { "score": 5, "user_vote": null }, "forecasts_count": 23, "key_factors": [], "is_current_content_translated": false, "description": "On October 21, 2022, Facebook, the social media powerhouse, rebranded to Meta Platforms, reflecting CEO Mark Zuckerberg's shift towards a virtual reality-focused future, termed the Metaverse. Following the rebranding, Meta's stock value dropped due to poor earnings, raising concerns about the firm's future profitability and relevance under Zuckerberg's new direction. While share prices have rebounded, ongoing financial struggles, challenges in making the Metaverse profitable, or doubts about the company's adaptability in the ever-changing social media landscape could potentially lead to the end of Zuckerberg's [leadership](https://about.meta.com/media-gallery/executives/) as CEO.\n\nSee also: \n\n- Washington Post: [With Bezos out as Amazon CEO, Zuckerberg is the last man standing](https://www.washingtonpost.com/technology/2021/02/03/bezos-zuckerberg-ceo-transition/)\n\n- Harvard Business Review: [Should Mark Zuckerberg Resign?](https://hbr.org/2019/01/should-mark-zuckerberg-resign)" }, { "id": 20241, "title": "Will Walmart stores in the United States return to 24-hour service before 2026?", "short_title": "Walmart open 24 hours again", "url_title": "Walmart open 24 hours again", "slug": "walmart-open-24-hours-again", "author_id": 130846, "author_username": "ForkLeaf", "coauthors": [], "created_at": "2023-11-29T08:40:05.717763Z", "published_at": "2023-12-10T17:30:00Z", "edited_at": "2025-09-21T22:47:49.360292Z", "curation_status": "approved", "curation_status_updated_at": "2023-12-10T17:30:00Z", "comment_count": 1, "status": "open", "resolved": false, "actual_close_time": null, "scheduled_close_time": "2026-01-01T04:00:00Z", "scheduled_resolve_time": "2026-01-01T05:00:00Z", "actual_resolve_time": null, "open_time": "2023-12-10T17:30:00Z", "nr_forecasters": 48, "html_metadata_json": null, "projects": { "leaderboard_tag": [ { "id": 32590, "name": "2021-2025 Leaderboard", "slug": "2021_2025_leaderboard", "type": "leaderboard_tag" } ], "site_main": [ { "id": 144, "type": "site_main", "name": "Metaculus Community", "slug": null, "header_image": null, "prize_pool": null, "start_date": null, "close_date": null, "forecasting_end_date": null, "html_metadata_json": null, "is_ongoing": null, "user_permission": "forecaster", "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2025-07-18T17:28:18.838588Z", "score_type": null, "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } ], "default_project": { "id": 144, "type": "site_main", "name": "Metaculus Community", "slug": null, "header_image": null, "prize_pool": null, "start_date": null, "close_date": null, "forecasting_end_date": null, "html_metadata_json": null, "is_ongoing": null, "user_permission": "forecaster", "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2025-07-18T17:28:18.838588Z", "score_type": null, "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" }, "category": [ { "id": 3698, "name": "Economy & Business", "slug": "economy-business", "emoji": "💼", "description": "Economy & Business", "type": "category" } ] }, "question": { "id": 20241, "title": "Will Walmart stores in the United States return to 24-hour service before 2026?", "created_at": "2023-11-29T08:40:05.717763Z", "open_time": "2023-12-10T17:30:00Z", "cp_reveal_time": "2023-12-13T17:30:00Z", "spot_scoring_time": "2023-12-13T17:30:00Z", "scheduled_resolve_time": "2026-01-01T05:00:00Z", "actual_resolve_time": null, "resolution_set_time": null, "scheduled_close_time": "2026-01-01T04:00:00Z", "actual_close_time": "2026-01-01T04:00:00Z", "type": "binary", "options": null, "group_variable": "", "status": "open", "possibilities": { "type": "binary" }, "resolution": null, "include_bots_in_aggregates": false, "question_weight": 1.0, "default_score_type": "peer", "default_aggregation_method": "recency_weighted", "label": "", "unit": "", "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "scaling": { "range_min": null, "range_max": null, "nominal_min": null, "nominal_max": null, "zero_point": null, "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "continuous_range": null }, "group_rank": null, "description": "[Walmart](https://en.wikipedia.org/wiki/Walmart) is a multinational retail store chain headquartered in the United States. Prior to 2020, Walmart stores were known for being open 24 hours a day, 7 days a week in most locations. However, COVID-19 and related government restrictions resulted in the company's management making the decision to close stores at night. In August of 2022, the chain announced that they had [no plans](https://www.wsfa.com/2022/08/12/walmart-no-plans-make-stores-247-again/) to make its stores open 24/7 again, likely as a result of [labor shortages](https://www.uschamber.com/workforce/understanding-americas-labor-shortage-the-most-impacted-industries). However, this has not stopped [speculation](https://www.cheddarflow.com/blog/the-end-of-24-hour-walmart-what-happened/) that the chain may return to 24-hour service at some point as labor shortages recede.", "resolution_criteria": "This question will resolve as **Yes** if, before January 1, 2026, [credible sources](https://www.metaculus.com/help/faq/#definitions) report that the majority of Walmart stores in the United States have returned to 24/7 service. The chain merely announcing that this is occurring will not be sufficient to resolve the question. If this does not occur before the resolution date, the question resolves as **No**.", "fine_print": "", "post_id": 20241, "aggregations": { "recency_weighted": { "history": [ { "start_time": 1758491561.873139, "end_time": 1763164608.237, "forecaster_count": 5, "interval_lower_bounds": [ 0.04 ], "centers": [ 0.04 ], "interval_upper_bounds": [ 0.1 ] } ], "latest": { "start_time": 1758491561.873139, "end_time": 1763164608.237, "forecaster_count": 5, "interval_lower_bounds": [ 0.04 ], "centers": [ 0.04 ], "interval_upper_bounds": [ 0.1 ], "forecast_values": [ 0.96, 0.04 ], "means": [ 0.0787437252539603 ], "histogram": [ [ 0.0, 0.0, 0.0, 0.0, 1.78972698844193, 0.0, 0.0, 0.0, 0.0, 0.0, 0.6040990104285446, 0.0, 0.2905243231860232, 0.0, 0.0, 0.0, 0.0, 0.0, 0.43961566820611525, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 ] ] }, "score_data": {}, "movement": null }, "unweighted": { "history": [], "latest": null, "score_data": {}, "movement": null }, "single_aggregation": { "history": [], "latest": null, "score_data": {}, "movement": null }, "metaculus_prediction": { "history": [ { "start_time": 1728289329.390586, "end_time": null, "forecaster_count": 14, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null } ], "latest": { "start_time": 1728289329.390586, "end_time": null, "forecaster_count": 14, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null, "forecast_values": [ 0.8666647680338668, 0.13333523196613312 ], "means": null, "histogram": null }, "score_data": {}, "movement": null } } }, "user_permission": "forecaster", "vote": { "score": 5, "user_vote": null }, "forecasts_count": 110, "key_factors": [], "is_current_content_translated": false, "description": "[Walmart](https://en.wikipedia.org/wiki/Walmart) is a multinational retail store chain headquartered in the United States. Prior to 2020, Walmart stores were known for being open 24 hours a day, 7 days a week in most locations. However, COVID-19 and related government restrictions resulted in the company's management making the decision to close stores at night. In August of 2022, the chain announced that they had [no plans](https://www.wsfa.com/2022/08/12/walmart-no-plans-make-stores-247-again/) to make its stores open 24/7 again, likely as a result of [labor shortages](https://www.uschamber.com/workforce/understanding-americas-labor-shortage-the-most-impacted-industries). However, this has not stopped [speculation](https://www.cheddarflow.com/blog/the-end-of-24-hour-walmart-what-happened/) that the chain may return to 24-hour service at some point as labor shortages recede." }, { "id": 20218, "title": "Will there be a white Christmas in at least 4 of these 9 large European cities in 2023?", "short_title": "White 2023 Christmas in Europe?", "url_title": "White 2023 Christmas in Europe?", "slug": "white-2023-christmas-in-europe", "author_id": 118622, "author_username": "kqr", "coauthors": [], "created_at": "2023-11-27T11:39:20.101935Z", "published_at": "2023-12-06T16:00:00Z", "edited_at": "2025-09-05T17:28:47.005919Z", "curation_status": "approved", "curation_status_updated_at": "2023-12-06T16:00:00Z", "comment_count": 34, "status": "resolved", "resolved": true, "actual_close_time": "2023-12-16T11:00:00Z", "scheduled_close_time": "2023-12-16T11:00:00Z", "scheduled_resolve_time": "2024-01-03T18:07:00Z", "actual_resolve_time": "2024-01-03T18:07:00Z", "open_time": "2023-12-06T16:00:00Z", "nr_forecasters": 47, "html_metadata_json": null, "projects": { "leaderboard_tag": [ { "id": 32591, "name": "2023 Leaderboard", "slug": "2023_leaderboard", "type": "leaderboard_tag" } ], "topic": [ { "id": 15867, "name": "Environment & Climate", "slug": "climate", "emoji": "🌎", "type": "topic" } ], "site_main": [ { "id": 144, "type": "site_main", "name": "Metaculus Community", "slug": null, "header_image": null, "prize_pool": null, "start_date": null, "close_date": null, "forecasting_end_date": null, "html_metadata_json": null, "is_ongoing": null, "user_permission": "forecaster", "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2025-07-18T17:28:18.838588Z", "score_type": null, "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } ], "category": [ { "id": 3697, "name": "Environment & Climate", "slug": "environment-climate", "emoji": "🌱", "description": "Environment & Climate", "type": "category" } ], "tournament": [ { "id": 2654, "type": "tournament", "name": "🏆 Q4 2023 Quarterly Cup 🏆", "slug": "quarterly-cup-2023q4", "header_image": "https://metaculus-web-media.s3.amazonaws.com/quarterlycup.webp", "prize_pool": null, "start_date": "2023-10-03T15:00:00Z", "close_date": "2024-01-04T12:00:00Z", "forecasting_end_date": "2023-12-31T23:00:00Z", "html_metadata_json": null, "is_ongoing": false, "user_permission": null, "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2024-01-05T15:54:55.645171Z", "score_type": "relative_legacy_tournament", "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } ], "default_project": { "id": 2654, "type": "tournament", "name": "🏆 Q4 2023 Quarterly Cup 🏆", "slug": "quarterly-cup-2023q4", "header_image": "https://metaculus-web-media.s3.amazonaws.com/quarterlycup.webp", "prize_pool": null, "start_date": "2023-10-03T15:00:00Z", "close_date": "2024-01-04T12:00:00Z", "forecasting_end_date": "2023-12-31T23:00:00Z", "html_metadata_json": null, "is_ongoing": false, "user_permission": null, "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2024-01-05T15:54:55.645171Z", "score_type": "relative_legacy_tournament", "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } }, "question": { "id": 20218, "title": "Will there be a white Christmas in at least 4 of these 9 large European cities in 2023?", "created_at": "2023-11-27T11:39:20.101935Z", "open_time": "2023-12-06T16:00:00Z", "cp_reveal_time": "2023-12-07T15:30:00Z", "spot_scoring_time": "2023-12-07T15:30:00Z", "scheduled_resolve_time": "2024-01-03T18:07:00Z", "actual_resolve_time": "2024-01-03T18:07:00Z", "resolution_set_time": "2024-01-03T18:07:00Z", "scheduled_close_time": "2023-12-16T11:00:00Z", "actual_close_time": "2023-12-16T11:00:00Z", "type": "binary", "options": null, "group_variable": "", "status": "resolved", "possibilities": { "type": "binary" }, "resolution": "no", "include_bots_in_aggregates": false, "question_weight": 1.0, "default_score_type": "peer", "default_aggregation_method": "recency_weighted", "label": "", "unit": "", "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "scaling": { "range_min": null, "range_max": null, "nominal_min": null, "nominal_max": null, "zero_point": null, "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "continuous_range": null }, "group_rank": null, "description": "Many children wish for a white Christmas, i.e. snow on the days 24th or 25th December. Transit operators and homeless people enjoy it less. It is an important cultural event around the northern Atlantic, though; hence this question!\n\nCultural note: The majority of inhabitants of Instanbul do not celebrate Christmas at all. Neither do all Russians celebrate Christmas in December. These differences are irrelevant for the question, which is focused on Christmas as celebrated by most cultures around the northern Atlantic.\n\n-----\n\nThis question uses METAR data as the source for convenience, because it provides a standardised view into actual observed weather at diverse geographical locations. Here's an example of a METAR report:\n\n> METAR UUEE 050800Z 28003MPS 240V320 7000 -SN SCT006 M11/M13 Q1018 R24L/490336 R24C/490336 NOSIG=\n\nThe standard allows for many optional components, but common components are\n\n- The reporting station (in this case `UUEE`, which is a Moscow airport) comes after the preamble\n- the day and time of observation (in this case `050800Z` meaning day 5 of the month, at 8 o'clock UTC)\n- wind speed and direction (in this case `28003MPS` meaning 280°, 3 m/s)\n- visibility (in this case `7000` meaning 7000 metres)\n- cloud cover (in this case `SCT006` meaning scattered at 600 feet)\n- temperature (in this case `M11/M13` meaning temperature of -11 °C)\n- air pressure (in this case `Q1018` meaning 1018 millibar)\n- runway information, starting with R\n- a near-term forecast (in this case `NOSIG` meaning no significant change expected)\n\nBut! Critically for this question, there are also indications of precipitation. In the example above, that's \"-SN\" for \"light snow\".\n\nHere are some other examples of reports indicating snow:\n\n> METAR ESSA 270120Z 01005KT 9999 -SN SCT007 BKN009 OVC046 M05/M06 Q1010 RESN BECMG BKN020=`\n\n`RESN` means \"recent snow\"\n\n> METAR ESNU 252220Z 33009KT 9999 SG FEW018 OVC055 M03/M05 Q1003=\n\n`SG` means \"snow grains\"\n\n> METAR EFOU 041820Z AUTO 14003KT 9999 -SHSN SCT044 M13/M15 Q1013=\n\n`-SHSN` means \"light snow shower\"\n\nFor decoding reports, [the web tool by Flight Plan Database](https://flightplandatabase.com/METAR) can be a helpful assistant.", "resolution_criteria": "The 10 largest cities in Europe [according to Wikipedia](https://en.wikipedia.org/wiki/List_of_European_cities_by_population_within_city_limits#Largest_cities) are\n\n- Istanbul\n- Moscow\n- London\n- Saint Petersburg\n- Berlin\n- Madrid\n- Kyiv\n- Rome\n- Bucharest\n- Paris\n\nFor this question, we unfortunately have to exclude Kyiv due to the lack of standardised aerodrome weather reporting in the area.\n\nEach of the remaining 9 cities will have been considered to have had a white Christmas if, on either of the days 24 December or 25 December, there has been precipitation in the form of snow. The selection will be set to begin at hour 0 UTC on the 24th and end at hour 23 UTC on the 25th.\n\nIf four or more of them have had a white Christmas, then this question resolves as **Yes**. Otherwise, it resolves as **No**.", "fine_print": "To determine whether there has been precipitation in these cities, the [Ogimet METAR](http://www.ogimet.com/metars.phtml.en) historic data for the dates given above will be consulted. If that source is down, a similar one will be located. If none can be found with reasonable effort, the question resolves **Ambiguous**.\n\nThe lookup will be made for the following airports (corresponding to the cities listed above, excluding Kyiv): LTFM, UUEE, EGLL, ULLI, EDDB, LEMD, LIRF, LROP, LFPG. Any reference to snow (SN), snow grains (SG), snow pellets (SP or GS), or snow shower (SW) will count as there being precipitation in the form of snow.", "post_id": 20218, "aggregations": { "recency_weighted": { "history": [ { "start_time": 1702718035.220868, "end_time": null, "forecaster_count": 47, "interval_lower_bounds": [ 0.17 ], "centers": [ 0.2 ], "interval_upper_bounds": [ 0.35 ] } ], "latest": { "start_time": 1702718035.220868, "end_time": null, "forecaster_count": 47, "interval_lower_bounds": [ 0.17 ], "centers": [ 0.2 ], "interval_upper_bounds": [ 0.35 ], "forecast_values": [ 0.8, 0.2 ], "means": [ 0.25191100130272565 ], "histogram": [ [ 0.0, 0.0, 0.0, 0.0, 0.800591043524475, 0.0, 0.0, 0.0, 0.0, 0.0, 0.3403259577906623, 0.0, 0.0, 0.5763984674598623, 0.6873484555754474, 0.3726419108865464, 0.0, 0.6975902937211316, 0.968955588553565, 0.0, 3.1239527170650456, 0.0, 0.0, 0.0, 0.0, 0.5874172525367372, 0.0, 0.012201884674984505, 0.09222549710312164, 0.0, 0.6213416309859594, 0.004333234413825695, 0.0, 0.0, 0.0, 0.9583406036902323, 0.0, 0.0, 0.0, 0.0, 0.8629050061859964, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.6360167605213906, 0.0, 0.0, 0.0, 0.0, 0.7529282881015206, 0.0, 0.03365636844874199, 0.0, 0.0, 0.03600755180602183, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.05065733435182447, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0028636604341700815 ] ] }, "score_data": { "peer_score": 20.45109908963727, "coverage": 0.9929552599183213, "baseline_score": 47.99070834865969, "spot_peer_score": 41.047876783067224, "peer_archived_score": 20.45109908963727, "baseline_archived_score": 47.99070834865969, "spot_peer_archived_score": 41.047876783067224 }, "movement": null }, "unweighted": { "history": [], "latest": null, "score_data": {}, "movement": null }, "single_aggregation": { "history": [], "latest": null, "score_data": {}, "movement": null }, "metaculus_prediction": { "history": [ { "start_time": 1702718035.254343, "end_time": null, "forecaster_count": 47, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null } ], "latest": { "start_time": 1702718035.254343, "end_time": null, "forecaster_count": 47, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null, "forecast_values": [ 0.9529730921289876, 0.047026907871012394 ], "means": null, "histogram": null }, "score_data": {}, "movement": null } } }, "user_permission": "forecaster", "vote": { "score": 6, "user_vote": null }, "forecasts_count": 174, "key_factors": [], "is_current_content_translated": false, "description": "Many children wish for a white Christmas, i.e. snow on the days 24th or 25th December. Transit operators and homeless people enjoy it less. It is an important cultural event around the northern Atlantic, though; hence this question!\n\nCultural note: The majority of inhabitants of Instanbul do not celebrate Christmas at all. Neither do all Russians celebrate Christmas in December. These differences are irrelevant for the question, which is focused on Christmas as celebrated by most cultures around the northern Atlantic.\n\n-----\n\nThis question uses METAR data as the source for convenience, because it provides a standardised view into actual observed weather at diverse geographical locations. Here's an example of a METAR report:\n\n> METAR UUEE 050800Z 28003MPS 240V320 7000 -SN SCT006 M11/M13 Q1018 R24L/490336 R24C/490336 NOSIG=\n\nThe standard allows for many optional components, but common components are\n\n- The reporting station (in this case `UUEE`, which is a Moscow airport) comes after the preamble\n- the day and time of observation (in this case `050800Z` meaning day 5 of the month, at 8 o'clock UTC)\n- wind speed and direction (in this case `28003MPS` meaning 280°, 3 m/s)\n- visibility (in this case `7000` meaning 7000 metres)\n- cloud cover (in this case `SCT006` meaning scattered at 600 feet)\n- temperature (in this case `M11/M13` meaning temperature of -11 °C)\n- air pressure (in this case `Q1018` meaning 1018 millibar)\n- runway information, starting with R\n- a near-term forecast (in this case `NOSIG` meaning no significant change expected)\n\nBut! Critically for this question, there are also indications of precipitation. In the example above, that's \"-SN\" for \"light snow\".\n\nHere are some other examples of reports indicating snow:\n\n> METAR ESSA 270120Z 01005KT 9999 -SN SCT007 BKN009 OVC046 M05/M06 Q1010 RESN BECMG BKN020=`\n\n`RESN` means \"recent snow\"\n\n> METAR ESNU 252220Z 33009KT 9999 SG FEW018 OVC055 M03/M05 Q1003=\n\n`SG` means \"snow grains\"\n\n> METAR EFOU 041820Z AUTO 14003KT 9999 -SHSN SCT044 M13/M15 Q1013=\n\n`-SHSN` means \"light snow shower\"\n\nFor decoding reports, [the web tool by Flight Plan Database](https://flightplandatabase.com/METAR) can be a helpful assistant." }, { "id": 20172, "title": "Before 2026, will OpenAI's commercial operations cease to be governed by its nonprofit board of directors?", "short_title": "OpenAI Cease Governance by Nonprofit Board?", "url_title": "OpenAI Cease Governance by Nonprofit Board?", "slug": "openai-cease-governance-by-nonprofit-board", "author_id": 127582, "author_username": "lbiii", "coauthors": [], "created_at": "2023-11-24T17:39:49.211714Z", "published_at": "2024-01-13T15:58:00Z", "edited_at": "2025-09-21T22:54:44.320250Z", "curation_status": "approved", "curation_status_updated_at": "2024-01-13T15:58:00Z", "comment_count": 18, "status": "open", "resolved": false, "actual_close_time": null, "scheduled_close_time": "2026-01-01T08:00:00Z", "scheduled_resolve_time": "2026-01-01T08:00:00Z", "actual_resolve_time": null, "open_time": "2024-01-13T15:58:00Z", "nr_forecasters": 104, "html_metadata_json": null, "projects": { "leaderboard_tag": [ { "id": 32601, "name": "2024-2025 Leaderboard", "slug": "2024_2025_leaderboard", "type": "leaderboard_tag" } ], "topic": [ { "id": 15869, "name": "Artificial Intelligence", "slug": "ai", "emoji": "🤖", "type": "topic" } ], "site_main": [ { "id": 144, "type": "site_main", "name": "Metaculus Community", "slug": null, "header_image": null, "prize_pool": null, "start_date": null, "close_date": null, "forecasting_end_date": null, "html_metadata_json": null, "is_ongoing": null, "user_permission": "forecaster", "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2025-07-18T17:28:18.838588Z", "score_type": null, "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } ], "default_project": { "id": 144, "type": "site_main", "name": "Metaculus Community", "slug": null, "header_image": null, "prize_pool": null, "start_date": null, "close_date": null, "forecasting_end_date": null, "html_metadata_json": null, "is_ongoing": null, "user_permission": "forecaster", "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2025-07-18T17:28:18.838588Z", "score_type": null, "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" }, "category": [ { "id": 3698, "name": "Economy & Business", "slug": "economy-business", "emoji": "💼", "description": "Economy & Business", "type": "category" }, { "id": 3692, "name": "Computing and Math", "slug": "computing-and-math", "emoji": "💻", "description": "Computing and Math", "type": "category" }, { "id": 3694, "name": "Artificial Intelligence", "slug": "artificial-intelligence", "emoji": "🤖", "description": "Artificial Intelligence", "type": "category" } ] }, "question": { "id": 20172, "title": "Before 2026, will OpenAI's commercial operations cease to be governed by its nonprofit board of directors?", "created_at": "2023-11-24T17:39:49.211714Z", "open_time": "2024-01-13T15:58:00Z", "cp_reveal_time": "2024-01-16T15:58:00Z", "spot_scoring_time": "2024-01-16T15:58:00Z", "scheduled_resolve_time": "2026-01-01T08:00:00Z", "actual_resolve_time": null, "resolution_set_time": null, "scheduled_close_time": "2026-01-01T08:00:00Z", "actual_close_time": "2026-01-01T08:00:00Z", "type": "binary", "options": null, "group_variable": "", "status": "open", "possibilities": { "type": "binary" }, "resolution": null, "include_bots_in_aggregates": false, "question_weight": 1.0, "default_score_type": "peer", "default_aggregation_method": "recency_weighted", "label": "", "unit": "", "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "scaling": { "range_min": null, "range_max": null, "nominal_min": null, "nominal_max": null, "zero_point": null, "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "continuous_range": null }, "group_rank": null, "description": "OpenAI Inc., a Delaware 501(c)3 nonprofit corporation, was founded in late 2015. At its [inception](https://web.archive.org/web/20151222103150/https://openai.com/blog/introducing-openai/), OpenAI's mission and corporate structure were intended to maximize and broadly distribute the benefits of AI for humanity while minimizing the risks of such a potentially transformative—and possibly harmful—technology.\n\nThe [tumultuous events](https://en.wikipedia.org/wiki/OpenAI#2023%E2%80%93present:_Brief_departure_of_Altman_and_Brockman) of November 17-21, 2023 culminated in a [reconstitution](https://www.cnbc.com/2023/11/22/sam-altmans-back-heres-whos-on-the-new-openai-board-and-whos-out.html) of the nonprofit's board of directors and, along with it, [speculation](https://www.nytimes.com/2023/11/22/technology/openai-board-capitalists.html) regarding whether the company's original mission and structure will remain intact.", "resolution_criteria": "This question resolves \"Yes\" if, before January 1, 2026, [credible sources](https://www.metaculus.com/help/faq/#definitions) report that OpenAI's commercial operations (such operations, as of the writing of this question, are housed under OpenAI Global LLC) are no longer governed by the nonprofit board of directors of OpenAI Inc.\n\n* **Resolution in the Event of Restructuring, Merger, etc.**—If, through corporate restructuring, merger, acquisition, transfer of assets (including intellectual property) or by other means, all or a substantial portion of OpenAI's commercial operations are transferred to another entity that is not controlled by the nonprofit board of directors of OpenAI, this question would resolve \"Yes\".\n\n* **Resolution in the Event of a Mass Departure of Staff to a Competitor or New Entity**—If within a 180-day period, more than 50% of employees of OpenAI Global LLC (or a successor entity that is still under the control of the nonprofit board) housing OpenAI's commercial activities become employees of another single entity that is not controlled by the nonprofit board of OpenAI Inc., this would be considered a substantial transfer of OpenAI's intellectual property, and this question would resolve \"Yes\".", "fine_print": "", "post_id": 20172, "aggregations": { "recency_weighted": { "history": [ { "start_time": 1758031496.625773, "end_time": 1759515979.570323, "forecaster_count": 30, "interval_lower_bounds": [ 0.01 ], "centers": [ 0.039 ], "interval_upper_bounds": [ 0.06 ] } ], "latest": { "start_time": 1758031496.625773, "end_time": 1759515979.570323, "forecaster_count": 30, "interval_lower_bounds": [ 0.01 ], "centers": [ 0.039 ], "interval_upper_bounds": [ 0.06 ], "forecast_values": [ 0.961, 0.039 ], "means": [ 0.11480625978311454 ], "histogram": [ [ 1.5137562793032666, 1.8398013581295882, 0.38213595752647456, 1.0106482519780755, 1.7839727899461608, 0.26666414070049616, 0.4245262689949799, 0.0, 0.05892591724387544, 0.0, 0.6014438147875389, 0.0, 0.2581744010322857, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.7549731054070336, 0.0, 0.011364900518339544, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.5608812398816437 ] ] }, "score_data": {}, "movement": null }, "unweighted": { "history": [], "latest": null, "score_data": {}, "movement": null }, "single_aggregation": { "history": [], "latest": null, "score_data": {}, "movement": null }, "metaculus_prediction": { "history": [ { "start_time": 1728289403.46587, "end_time": null, "forecaster_count": 34, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null } ], "latest": { "start_time": 1728289403.46587, "end_time": null, "forecaster_count": 34, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null, "forecast_values": [ 0.17756488886418553, 0.8224351111358145 ], "means": null, "histogram": null }, "score_data": {}, "movement": null } } }, "user_permission": "forecaster", "vote": { "score": 7, "user_vote": null }, "forecasts_count": 292, "key_factors": [], "is_current_content_translated": false, "description": "OpenAI Inc., a Delaware 501(c)3 nonprofit corporation, was founded in late 2015. At its [inception](https://web.archive.org/web/20151222103150/https://openai.com/blog/introducing-openai/), OpenAI's mission and corporate structure were intended to maximize and broadly distribute the benefits of AI for humanity while minimizing the risks of such a potentially transformative—and possibly harmful—technology.\n\nThe [tumultuous events](https://en.wikipedia.org/wiki/OpenAI#2023%E2%80%93present:_Brief_departure_of_Altman_and_Brockman) of November 17-21, 2023 culminated in a [reconstitution](https://www.cnbc.com/2023/11/22/sam-altmans-back-heres-whos-on-the-new-openai-board-and-whos-out.html) of the nonprofit's board of directors and, along with it, [speculation](https://www.nytimes.com/2023/11/22/technology/openai-board-capitalists.html) regarding whether the company's original mission and structure will remain intact." }, { "id": 20171, "title": "Will Huada Empyrean be publicly traded in 2026?", "short_title": "Huada Empyrean public in 2026", "url_title": "Huada Empyrean public in 2026", "slug": "huada-empyrean-public-in-2026", "author_id": 111541, "author_username": "DerMannOhneEigenschaften", "coauthors": [], "created_at": "2023-11-24T16:13:37.081590Z", "published_at": "2023-12-06T11:00:00Z", "edited_at": "2025-09-05T17:28:58.070594Z", "curation_status": "approved", "curation_status_updated_at": "2023-12-06T11:00:00Z", "comment_count": 9, "status": "resolved", "resolved": true, "actual_close_time": "2024-01-29T14:14:00Z", "scheduled_close_time": "2026-01-01T11:00:00Z", "scheduled_resolve_time": "2026-01-01T11:00:00Z", "actual_resolve_time": "2024-01-29T14:14:00Z", "open_time": "2023-12-06T11:00:00Z", "nr_forecasters": 12, "html_metadata_json": null, "projects": { "leaderboard_tag": [ { "id": 32590, "name": "2021-2025 Leaderboard", "slug": "2021_2025_leaderboard", "type": "leaderboard_tag" } ], "site_main": [ { "id": 144, "type": "site_main", "name": "Metaculus Community", "slug": null, "header_image": null, "prize_pool": null, "start_date": null, "close_date": null, "forecasting_end_date": null, "html_metadata_json": null, "is_ongoing": null, "user_permission": "forecaster", "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2025-07-18T17:28:18.838588Z", "score_type": null, "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } ], "tournament": [ { "id": 2784, "type": "tournament", "name": "Chinese AI Chips", "slug": "chinese-ai-chips", "header_image": "https://metaculus-web-media.s3.amazonaws.com/cac-new_1.webp", "prize_pool": null, "start_date": "2023-12-06T12:00:33Z", "close_date": "2033-01-02T23:59:33Z", "forecasting_end_date": "2033-01-01T11:00:00Z", "html_metadata_json": null, "is_ongoing": true, "user_permission": null, "created_at": "2023-11-28T16:30:32.190353Z", "edited_at": "2025-09-22T00:55:08.932036Z", "score_type": "peer_tournament", "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } ], "default_project": { "id": 2784, "type": "tournament", "name": "Chinese AI Chips", "slug": "chinese-ai-chips", "header_image": "https://metaculus-web-media.s3.amazonaws.com/cac-new_1.webp", "prize_pool": null, "start_date": "2023-12-06T12:00:33Z", "close_date": "2033-01-02T23:59:33Z", "forecasting_end_date": "2033-01-01T11:00:00Z", "html_metadata_json": null, "is_ongoing": true, "user_permission": null, "created_at": "2023-11-28T16:30:32.190353Z", "edited_at": "2025-09-22T00:55:08.932036Z", "score_type": "peer_tournament", "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" }, "category": [ { "id": 3701, "name": "Technology", "slug": "technology", "emoji": "⚙️", "description": "Technology", "type": "category" }, { "id": 3692, "name": "Computing and Math", "slug": "computing-and-math", "emoji": "💻", "description": "Computing and Math", "type": "category" } ], "question_series": [ { "id": 2343, "type": "question_series", "name": "AI in China", "slug": "ai-in-china", "header_image": "https://metaculus-web-media.s3.amazonaws.com/ai-china.png", "prize_pool": null, "start_date": null, "close_date": null, "forecasting_end_date": null, "html_metadata_json": null, "is_ongoing": true, "user_permission": "forecaster", "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2025-09-22T00:55:09.020170Z", "score_type": "relative_legacy_tournament", "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } ] }, "question": { "id": 20171, "title": "Will Huada Empyrean be publicly traded in 2026?", "created_at": "2023-11-24T16:13:37.081590Z", "open_time": "2023-12-06T11:00:00Z", "cp_reveal_time": "2023-12-20T11:00:00Z", "spot_scoring_time": "2023-12-20T11:00:00Z", "scheduled_resolve_time": "2026-01-01T11:00:00Z", "actual_resolve_time": "2024-01-29T14:14:00Z", "resolution_set_time": "2024-01-29T14:14:00Z", "scheduled_close_time": "2026-01-01T11:00:00Z", "actual_close_time": "2024-01-29T14:14:00Z", "type": "binary", "options": null, "group_variable": "", "status": "resolved", "possibilities": { "type": "binary" }, "resolution": "annulled", "include_bots_in_aggregates": false, "question_weight": 1.0, "default_score_type": "peer", "default_aggregation_method": "recency_weighted", "label": "", "unit": "", "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "scaling": { "range_min": null, "range_max": null, "nominal_min": null, "nominal_max": null, "zero_point": null, "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "continuous_range": null }, "group_rank": null, "description": "The Chinese government is committed to indigenizing AI chip design and manufacture, having set it as a national priority and invested many billions of dollars in related ventures. While the Chinese semiconductor industry has had some notable successes in chip design and fabrication, it faces the formidable challenge of trying to replicate a highly advanced and globalized supply chain – currently spread out over Germany, Japan, the Netherlands, South Korea, Taiwan, the United States, and elsewhere – in one country alone.\n\nChip design has a number of critical dependencies. Most notably, it relies on electronic design automation (EDA) software for translating high-level designs into circuit layouts, and for simulating and validating chip architectures before they are fabricated. The EDA industry is dominated by three companies: Cadence (US), Synopsys (US), and Siemens EDA (US-based but German-owned). Chip designers also make use of core IP, pre-designed and verified components that can be licensed from vendors and integrated into designs. Core IP is offered as part of EDA suites, and also by dedicated core IP firms, the most prominent of which is Arm (UK). Chinese AI chip designers likely rely on foreign EDA software and core IP, both of which can, in some circumstances at least, be pirated.\n\nThe Chinese EDA industry is [small but growing](https://perma.cc/V6EN-U48B). Notable companies offering EDA suites include Empyrean Technology and Huawei. Chinese EDA firms have received major subsidies, and some have seen sales [grow by 30%](https://perma.cc/A8WC-MRL6) in the first three quarters of 2023. Empyrean [claims](https://perma.cc/H4L2-U3QM) that its EDA suite supports designs for chips at 7 nm, and Huawei [claims](https://perma.cc/6E4A-S7EA) its software supports designs for chips at 14 nm and above. However, Cadence, Synopsys, and Siemens EDA [still combine](https://perma.cc/A8WC-MRL6) for more than 80% of the Chinese market. To comply with [export controls](https://www.ecfr.gov/current/title-15/subtitle-B/chapter-VII/subchapter-C), American EDA firms must customize their tools to exclude support for gate-all-around field-effect transistor (GAAFET) designs when sold to Chinese customers.\n\nEmpyrean is currently private, but if it became listed on a public stock exchange, that would be an indication that it is doing well.\n\nFor more information, I have written a primer on [indigenously made Chinese AI chips](https://docs.google.com/document/d/1r-aLT67d2OQnuhZJZihkr0KxxB-PsDMKMut2flaYQ2c/edit?usp=sharing), intended to provide a useful introduction and reference for anyone interested in forecasting or learning more about China’s ability to indigenously produce AI chips. It explains some key concepts, provides an overview of the chip-making process, provides an overview of relevant export controls, and describes key inputs, context, and organizations relevant to understanding Chinese chip-making progress.", "resolution_criteria": "The question will resolve as **Yes** if [Huada Empyrean](https://finance.yahoo.com/company/huada-empyrean-software?h=eyJlIjoiaHVhZGEtZW1weXJlYW4tc29mdHdhcmUiLCJuIjoiSHVhZGEgRW1weXJlYW4gU29mdHdhcmUifQ==&.tsrc=fin-srch) is listed on a public stock exchange on January 1st, 2026. If not listed on that date, the question will resolve as **No**.", "fine_print": "A “public stock exchange” includes but is not limited to the Shanghai Stock Exchange, the Shenzhen Stock Exchange, the Hong Kong Stock Exchange, Nasdaq, and the New York Stock Exchange; we accept any regulated marketplace where company shares are publicly issued and traded.\n\nIf the company changes its name, it is still considered the same company, if and only if the name change can be confirmed via Wikipedia or credible news reports.\n\nIf the company is acquired by another company or undergoes a merger with another company, the question will be **annulled**. (Note that this excludes [special-purpose acquisition companies](https://en.wikipedia.org/wiki/Special-purpose_acquisition_company) (SPACs)).", "post_id": 20171, "aggregations": { "recency_weighted": { "history": [ { "start_time": 1706280539.854825, "end_time": null, "forecaster_count": 12, "interval_lower_bounds": [ 0.36 ], "centers": [ 0.45 ], "interval_upper_bounds": [ 0.52 ] } ], "latest": { "start_time": 1706280539.854825, "end_time": null, "forecaster_count": 12, "interval_lower_bounds": [ 0.36 ], "centers": [ 0.45 ], "interval_upper_bounds": [ 0.52 ], "forecast_values": [ 0.55, 0.45 ], "means": [ 0.45253155762782327 ], "histogram": [ [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.7064993787569234, 0.0, 0.0, 0.0, 0.0, 0.6554110117958545, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.4411588324588233, 0.0, 0.12874931591104752, 0.0, 0.0, 0.0, 0.0, 0.0, 0.08508524734423982, 0.862882435577901, 0.0, 0.0, 0.0, 0.0, 1.0, 0.0, 0.7394682331392083, 0.0, 0.0, 0.23128568172579037, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.6286996658179782, 0.0, 0.0, 0.0, 0.0 ] ] }, "score_data": {}, "movement": null }, "unweighted": { "history": [], "latest": null, "score_data": {}, "movement": null }, "single_aggregation": { "history": [], "latest": null, "score_data": {}, "movement": null }, "metaculus_prediction": { "history": [ { "start_time": 1706514482.978066, "end_time": null, "forecaster_count": 12, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null } ], "latest": { "start_time": 1706514482.978066, "end_time": null, "forecaster_count": 12, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null, "forecast_values": [ 0.6678387851182983, 0.33216121488170175 ], "means": null, "histogram": null }, "score_data": {}, "movement": null } } }, "user_permission": "forecaster", "vote": { "score": 5, "user_vote": null }, "forecasts_count": 47, "key_factors": [], "is_current_content_translated": false, "description": "The Chinese government is committed to indigenizing AI chip design and manufacture, having set it as a national priority and invested many billions of dollars in related ventures. While the Chinese semiconductor industry has had some notable successes in chip design and fabrication, it faces the formidable challenge of trying to replicate a highly advanced and globalized supply chain – currently spread out over Germany, Japan, the Netherlands, South Korea, Taiwan, the United States, and elsewhere – in one country alone.\n\nChip design has a number of critical dependencies. Most notably, it relies on electronic design automation (EDA) software for translating high-level designs into circuit layouts, and for simulating and validating chip architectures before they are fabricated. The EDA industry is dominated by three companies: Cadence (US), Synopsys (US), and Siemens EDA (US-based but German-owned). Chip designers also make use of core IP, pre-designed and verified components that can be licensed from vendors and integrated into designs. Core IP is offered as part of EDA suites, and also by dedicated core IP firms, the most prominent of which is Arm (UK). Chinese AI chip designers likely rely on foreign EDA software and core IP, both of which can, in some circumstances at least, be pirated.\n\nThe Chinese EDA industry is [small but growing](https://perma.cc/V6EN-U48B). Notable companies offering EDA suites include Empyrean Technology and Huawei. Chinese EDA firms have received major subsidies, and some have seen sales [grow by 30%](https://perma.cc/A8WC-MRL6) in the first three quarters of 2023. Empyrean [claims](https://perma.cc/H4L2-U3QM) that its EDA suite supports designs for chips at 7 nm, and Huawei [claims](https://perma.cc/6E4A-S7EA) its software supports designs for chips at 14 nm and above. However, Cadence, Synopsys, and Siemens EDA [still combine](https://perma.cc/A8WC-MRL6) for more than 80% of the Chinese market. To comply with [export controls](https://www.ecfr.gov/current/title-15/subtitle-B/chapter-VII/subchapter-C), American EDA firms must customize their tools to exclude support for gate-all-around field-effect transistor (GAAFET) designs when sold to Chinese customers.\n\nEmpyrean is currently private, but if it became listed on a public stock exchange, that would be an indication that it is doing well.\n\nFor more information, I have written a primer on [indigenously made Chinese AI chips](https://docs.google.com/document/d/1r-aLT67d2OQnuhZJZihkr0KxxB-PsDMKMut2flaYQ2c/edit?usp=sharing), intended to provide a useful introduction and reference for anyone interested in forecasting or learning more about China’s ability to indigenously produce AI chips. It explains some key concepts, provides an overview of the chip-making process, provides an overview of relevant export controls, and describes key inputs, context, and organizations relevant to understanding Chinese chip-making progress." }, { "id": 20169, "title": "Will a Chinese fab achieve a process with a density of greater than 150M transistors per square millimeter by 2027?", "short_title": "Chinese fab with >150 MTr/mm² process by 2027", "url_title": "Chinese fab with >150 MTr/mm² process by 2027", "slug": "chinese-fab-with-150-mtrmm2-process-by-2027", "author_id": 111541, "author_username": "DerMannOhneEigenschaften", "coauthors": [], "created_at": "2023-11-24T14:49:56.459591Z", "published_at": "2023-12-06T11:00:00Z", "edited_at": "2025-09-14T18:53:21.789203Z", "curation_status": "approved", "curation_status_updated_at": "2023-12-06T11:00:00Z", "comment_count": 0, "status": "open", "resolved": false, "actual_close_time": null, "scheduled_close_time": "2027-01-01T11:00:00Z", "scheduled_resolve_time": "2027-01-01T11:00:00Z", "actual_resolve_time": null, "open_time": "2023-12-06T11:00:00Z", "nr_forecasters": 31, "html_metadata_json": null, "projects": { "site_main": [ { "id": 144, "type": "site_main", "name": "Metaculus Community", "slug": null, "header_image": null, "prize_pool": null, "start_date": null, "close_date": null, "forecasting_end_date": null, "html_metadata_json": null, "is_ongoing": null, "user_permission": "forecaster", "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2025-07-18T17:28:18.838588Z", "score_type": null, "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } ], "tournament": [ { "id": 2784, "type": "tournament", "name": "Chinese AI Chips", "slug": "chinese-ai-chips", "header_image": "https://metaculus-web-media.s3.amazonaws.com/cac-new_1.webp", "prize_pool": null, "start_date": "2023-12-06T12:00:33Z", "close_date": "2033-01-02T23:59:33Z", "forecasting_end_date": "2033-01-01T11:00:00Z", "html_metadata_json": null, "is_ongoing": true, "user_permission": null, "created_at": "2023-11-28T16:30:32.190353Z", "edited_at": "2025-09-22T00:55:08.932036Z", "score_type": "peer_tournament", "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } ], "default_project": { "id": 2784, "type": "tournament", "name": "Chinese AI Chips", "slug": "chinese-ai-chips", "header_image": "https://metaculus-web-media.s3.amazonaws.com/cac-new_1.webp", "prize_pool": null, "start_date": "2023-12-06T12:00:33Z", "close_date": "2033-01-02T23:59:33Z", "forecasting_end_date": "2033-01-01T11:00:00Z", "html_metadata_json": null, "is_ongoing": true, "user_permission": null, "created_at": "2023-11-28T16:30:32.190353Z", "edited_at": "2025-09-22T00:55:08.932036Z", "score_type": "peer_tournament", "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" }, "category": [ { "id": 3701, "name": "Technology", "slug": "technology", "emoji": "⚙️", "description": "Technology", "type": "category" }, { "id": 3692, "name": "Computing and Math", "slug": "computing-and-math", "emoji": "💻", "description": "Computing and Math", "type": "category" } ], "question_series": [ { "id": 2343, "type": "question_series", "name": "AI in China", "slug": "ai-in-china", "header_image": "https://metaculus-web-media.s3.amazonaws.com/ai-china.png", "prize_pool": null, "start_date": null, "close_date": null, "forecasting_end_date": null, "html_metadata_json": null, "is_ongoing": true, "user_permission": "forecaster", "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2025-09-22T00:55:09.020170Z", "score_type": "relative_legacy_tournament", "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } ] }, "question": { "id": 20169, "title": "Will a Chinese fab achieve a process with a density of greater than 150M transistors per square millimeter by 2027?", "created_at": "2023-11-24T14:49:56.459591Z", "open_time": "2023-12-06T11:00:00Z", "cp_reveal_time": "2023-12-20T11:00:00Z", "spot_scoring_time": "2023-12-20T11:00:00Z", "scheduled_resolve_time": "2027-01-01T11:00:00Z", "actual_resolve_time": null, "resolution_set_time": null, "scheduled_close_time": "2027-01-01T11:00:00Z", "actual_close_time": "2027-01-01T11:00:00Z", "type": "binary", "options": null, "group_variable": "", "status": "open", "possibilities": { "type": "binary" }, "resolution": null, "include_bots_in_aggregates": false, "question_weight": 1.0, "default_score_type": "peer", "default_aggregation_method": "recency_weighted", "label": "", "unit": "", "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "scaling": { "range_min": null, "range_max": null, "nominal_min": null, "nominal_max": null, "zero_point": null, "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "continuous_range": null }, "group_rank": null, "description": "In October 2022, the US instituted [new, wide-ranging export controls](https://www.csis.org/analysis/choking-chinas-access-future-ai) aimed at controlling a set of “chokepoint” technologies in the global semiconductor supply chain. China is one of the leading countries in AI, but AI advancements have historically run on compute, and China’s access to cost-competitive machine learning (ML) compute -- the Chinese AI industry currently relies in large part on AI chips designed and fabricated abroad, and in particular Nvidia GPUs -- is now in question.\n\nThe [US export controls](https://www.ecfr.gov/current/title-15/subtitle-B/chapter-VII/subchapter-C) prevent the export of high-performance AI chips to China, as well as the export of advanced tooling and materials used to make such chips. Still, one way for China to gain better access to cost-competitive ML compute is to indigenously produce high-quality AI chips. In 2015, China set out an ambitious plan aiming to bring the proportion of chips bought by Chinese firms that is domestically produced from 15% to 70% by 2025; as of 2021 [that number had gone up to 24%](https://asia.nikkei.com/Politics/International-relations/US-China-tensions/U.S.-tech-curbs-threaten-China-s-quest-for-chip-independence). With the October restrictions -- and with US allies like Taiwan, Japan, South Korea, and the Netherlands enacting their own controls -- Chinese firms now face substantial challenges in their attempts to indigenize semiconductor production.\n\nChips are fabricated in “fabs” -- semiconductor fabrication plants where the chips are manufactured on silicon wafers. Notable Chinese semiconductor manufacturing companies include [Semiconductor Manufacturing International Corporation (SMIC)](https://finance.yahoo.com/quote/0981.HK?p=0981.HK&.tsrc=fin-srch) and [Hua Hong Semiconductor Limited](https://finance.yahoo.com/quote/1347.HK?p=1347.HK&.tsrc=fin-srch).\n\nIn August 2023, Huawei launched a new smartphone [containing a chip that was fabricated on SMIC's 7 nm process](https://www.csis.org/analysis/chip-race-china-gives-huawei-steering-wheel-huaweis-new-smartphone-and-future), albeit with foreign-made equipment and materials. This process could fit [an estimated](https://en.wikipedia.org/wiki/7_nm_process) 113.6 million transistors per square millimeter (MTr/mm²). The most advanced process today is TSMC’s N3E, belonging to the 3 nm class, which can fit [an estimated](https://fuse.wikichip.org/news/7375/tsmc-n3-and-challenges-ahead/) 216 MTr/mm². (TSMC reached 150 MTr/mm² around 2021, and Samsung around 2022.)\n\nFor more information, I have written a primer on [indigenously made Chinese AI chips](https://docs.google.com/document/d/1r-aLT67d2OQnuhZJZihkr0KxxB-PsDMKMut2flaYQ2c/edit?usp=sharing), intended to provide a useful introduction and reference for anyone interested in forecasting or learning more about China’s ability to indigenously produce AI chips. It explains some key concepts, provides an overview of the chip-making process, provides an overview of relevant export controls, and describes key inputs, context, and organizations relevant to understanding Chinese chip-making progress.", "resolution_criteria": "The question will resolve as **Yes** if a Chinese fab has a process for a chip that is in production that is estimated to fit more than 150 MTr/mm², before January 1st, 2027. The question will resolve as **No** otherwise.\n\nIf the PRC annexes Taiwan, then annexed-Taiwan counts as being part of China (i.e., fabs in annexed-Taiwan are considered Chinese fabs).", "fine_print": "The question will be resolved by looking at [WikiChip](https://en.wikichip.org/wiki/3_nm_lithography_process) and [Wikipedia](https://en.wikipedia.org/wiki/3_nm_process) (not only the linked pages, but also pages for other process nodes) and taking the highest value for an in-production chip backed by a sourced estimate that appears for a Chinese fab, if density estimates are generally available. If density estimates are not generally available, the question will be **annulled**. A sourced estimate will be one appearing on the wiki page and linking to or otherwise backed by a credible external source. See the last section in the fine print for more details.\n\n\"In-production\" means the chip must have at least entered [\"risk-production\"](https://www.extremetech.com/computing/320460-apple-tsmc-on-track-to-move-3nm-into-risk-production-by-the-end-of-2021#:~:text=%22Risk%20production%22%20is,overall%20foundry%20timelines.).\n\nA “Chinese fab” is a semiconductor fabrication facility located in China and owned by a company that is headquartered in China. “China” includes mainland China, Hong Kong, and Macao (but not Taiwan).\n\n**Note:** The details described below for assessing the density estimates alongside the availability of information are to avoid having a component of the probabilities for this question represent the likelihood of relevant estimates being published. The procedure should result in the question being **annulled** if density estimates generally are not available for Chinese fabs, so that forecasters can just predict on the probability that a Chinese fab will achieve such a density without giving (much) consideration to whether estimates will be published.\n\nSpecifically, the following procedure will be used to assess the density estimates and their general availability:\n\n* Chinese fabs will be identified on the wiki pages for potentially qualifying process nodes.\n* If there are no Chinese fabs listed for the process nodes that could potentially have the required density this question will resolve as **No**.\n* Otherwise if a Chinese fab is listed, and one or more of the listed processes for the Chinese lab contains a sourced estimate for an in-production chip exceeding 150 MTr/mm², the question resolves as **Yes**.\n* Otherwise, if 50% or more of the processes for Chinese fabs listed under the potentially qualifying nodes have credible estimates available, and none of these contains a sourced estimate exceeding 150 MTr/mm², this question will resolve as **No**.\n* Otherwise, the question will be **annulled**.", "post_id": 20169, "aggregations": { "recency_weighted": { "history": [ { "start_time": 1758275867.619, "end_time": 1759780673.63166, "forecaster_count": 14, "interval_lower_bounds": [ 0.72 ], "centers": [ 0.79 ], "interval_upper_bounds": [ 0.86 ] } ], "latest": { "start_time": 1758275867.619, "end_time": 1759780673.63166, "forecaster_count": 14, "interval_lower_bounds": [ 0.72 ], "centers": [ 0.79 ], "interval_upper_bounds": [ 0.86 ], "forecast_values": [ 0.20999999999999996, 0.79 ], "means": [ 0.7609023904479625 ], "histogram": [ [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.560245764355437, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.33423662601949417, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.2396931538549426, 0.0, 0.6231125364833519, 0.0, 0.0, 0.27467474026865935, 0.47632380816539127, 0.0, 0.0, 1.8727500189273638, 0.0, 0.0, 0.09754477024200368, 0.0, 0.0, 0.0, 0.7576333089753814, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.7877898738940653 ] ] }, "score_data": {}, "movement": null }, "unweighted": { "history": [], "latest": null, "score_data": {}, "movement": null }, "single_aggregation": { "history": [], "latest": null, "score_data": {}, "movement": null }, "metaculus_prediction": { "history": [ { "start_time": 1728287937.453364, "end_time": null, "forecaster_count": 20, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null } ], "latest": { "start_time": 1728287937.453364, "end_time": null, "forecaster_count": 20, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null, "forecast_values": [ 0.35526706146471243, 0.6447329385352876 ], "means": null, "histogram": null }, "score_data": {}, "movement": null } } }, "user_permission": "forecaster", "vote": { "score": 10, "user_vote": null }, "forecasts_count": 105, "key_factors": [], "is_current_content_translated": false, "description": "In October 2022, the US instituted [new, wide-ranging export controls](https://www.csis.org/analysis/choking-chinas-access-future-ai) aimed at controlling a set of “chokepoint” technologies in the global semiconductor supply chain. China is one of the leading countries in AI, but AI advancements have historically run on compute, and China’s access to cost-competitive machine learning (ML) compute -- the Chinese AI industry currently relies in large part on AI chips designed and fabricated abroad, and in particular Nvidia GPUs -- is now in question.\n\nThe [US export controls](https://www.ecfr.gov/current/title-15/subtitle-B/chapter-VII/subchapter-C) prevent the export of high-performance AI chips to China, as well as the export of advanced tooling and materials used to make such chips. Still, one way for China to gain better access to cost-competitive ML compute is to indigenously produce high-quality AI chips. In 2015, China set out an ambitious plan aiming to bring the proportion of chips bought by Chinese firms that is domestically produced from 15% to 70% by 2025; as of 2021 [that number had gone up to 24%](https://asia.nikkei.com/Politics/International-relations/US-China-tensions/U.S.-tech-curbs-threaten-China-s-quest-for-chip-independence). With the October restrictions -- and with US allies like Taiwan, Japan, South Korea, and the Netherlands enacting their own controls -- Chinese firms now face substantial challenges in their attempts to indigenize semiconductor production.\n\nChips are fabricated in “fabs” -- semiconductor fabrication plants where the chips are manufactured on silicon wafers. Notable Chinese semiconductor manufacturing companies include [Semiconductor Manufacturing International Corporation (SMIC)](https://finance.yahoo.com/quote/0981.HK?p=0981.HK&.tsrc=fin-srch) and [Hua Hong Semiconductor Limited](https://finance.yahoo.com/quote/1347.HK?p=1347.HK&.tsrc=fin-srch).\n\nIn August 2023, Huawei launched a new smartphone [containing a chip that was fabricated on SMIC's 7 nm process](https://www.csis.org/analysis/chip-race-china-gives-huawei-steering-wheel-huaweis-new-smartphone-and-future), albeit with foreign-made equipment and materials. This process could fit [an estimated](https://en.wikipedia.org/wiki/7_nm_process) 113.6 million transistors per square millimeter (MTr/mm²). The most advanced process today is TSMC’s N3E, belonging to the 3 nm class, which can fit [an estimated](https://fuse.wikichip.org/news/7375/tsmc-n3-and-challenges-ahead/) 216 MTr/mm². (TSMC reached 150 MTr/mm² around 2021, and Samsung around 2022.)\n\nFor more information, I have written a primer on [indigenously made Chinese AI chips](https://docs.google.com/document/d/1r-aLT67d2OQnuhZJZihkr0KxxB-PsDMKMut2flaYQ2c/edit?usp=sharing), intended to provide a useful introduction and reference for anyone interested in forecasting or learning more about China’s ability to indigenously produce AI chips. It explains some key concepts, provides an overview of the chip-making process, provides an overview of relevant export controls, and describes key inputs, context, and organizations relevant to understanding Chinese chip-making progress." }, { "id": 20157, "title": "Will any Chinese semiconductor foundry have 20% or more global semiconductor market share before 2033?", "short_title": "Chinese foundry with ≥20% of semi market?", "url_title": "Chinese foundry with ≥20% of semi market?", "slug": "chinese-foundry-with-20-of-semi-market", "author_id": 111541, "author_username": "DerMannOhneEigenschaften", "coauthors": [], "created_at": "2023-11-24T13:33:56.558355Z", "published_at": "2023-12-06T11:00:00Z", "edited_at": "2025-09-22T00:55:08.548179Z", "curation_status": "approved", "curation_status_updated_at": "2023-12-06T11:00:00Z", "comment_count": 7, "status": "open", "resolved": false, "actual_close_time": null, "scheduled_close_time": "2033-01-01T11:00:00Z", "scheduled_resolve_time": "2033-03-01T11:00:00Z", "actual_resolve_time": null, "open_time": "2023-12-06T11:00:00Z", "nr_forecasters": 54, "html_metadata_json": null, "projects": { "site_main": [ { "id": 144, "type": "site_main", "name": "Metaculus Community", "slug": null, "header_image": null, "prize_pool": null, "start_date": null, "close_date": null, "forecasting_end_date": null, "html_metadata_json": null, "is_ongoing": null, "user_permission": "forecaster", "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2025-07-18T17:28:18.838588Z", "score_type": null, "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } ], "tournament": [ { "id": 2784, "type": "tournament", "name": "Chinese AI Chips", "slug": "chinese-ai-chips", "header_image": "https://metaculus-web-media.s3.amazonaws.com/cac-new_1.webp", "prize_pool": null, "start_date": "2023-12-06T12:00:33Z", "close_date": "2033-01-02T23:59:33Z", "forecasting_end_date": "2033-01-01T11:00:00Z", "html_metadata_json": null, "is_ongoing": true, "user_permission": null, "created_at": "2023-11-28T16:30:32.190353Z", "edited_at": "2025-09-22T00:55:08.932036Z", "score_type": "peer_tournament", "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } ], "default_project": { "id": 2784, "type": "tournament", "name": "Chinese AI Chips", "slug": "chinese-ai-chips", "header_image": "https://metaculus-web-media.s3.amazonaws.com/cac-new_1.webp", "prize_pool": null, "start_date": "2023-12-06T12:00:33Z", "close_date": "2033-01-02T23:59:33Z", "forecasting_end_date": "2033-01-01T11:00:00Z", "html_metadata_json": null, "is_ongoing": true, "user_permission": null, "created_at": "2023-11-28T16:30:32.190353Z", "edited_at": "2025-09-22T00:55:08.932036Z", "score_type": "peer_tournament", "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" }, "category": [ { "id": 3701, "name": "Technology", "slug": "technology", "emoji": "⚙️", "description": "Technology", "type": "category" }, { "id": 3692, "name": "Computing and Math", "slug": "computing-and-math", "emoji": "💻", "description": "Computing and Math", "type": "category" } ], "question_series": [ { "id": 2343, "type": "question_series", "name": "AI in China", "slug": "ai-in-china", "header_image": "https://metaculus-web-media.s3.amazonaws.com/ai-china.png", "prize_pool": null, "start_date": null, "close_date": null, "forecasting_end_date": null, "html_metadata_json": null, "is_ongoing": true, "user_permission": "forecaster", "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2025-09-22T00:55:09.020170Z", "score_type": "relative_legacy_tournament", "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } ] }, "question": { "id": 20157, "title": "Will any Chinese semiconductor foundry have 20% or more global semiconductor market share before 2033?", "created_at": "2023-11-24T13:33:56.558355Z", "open_time": "2023-12-06T11:00:00Z", "cp_reveal_time": "2023-12-20T11:00:00Z", "spot_scoring_time": "2023-12-20T11:00:00Z", "scheduled_resolve_time": "2033-03-01T11:00:00Z", "actual_resolve_time": null, "resolution_set_time": null, "scheduled_close_time": "2033-01-01T11:00:00Z", "actual_close_time": "2033-01-01T11:00:00Z", "type": "binary", "options": null, "group_variable": "", "status": "open", "possibilities": { "type": "binary" }, "resolution": null, "include_bots_in_aggregates": false, "question_weight": 1.0, "default_score_type": "peer", "default_aggregation_method": "recency_weighted", "label": "", "unit": "", "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "scaling": { "range_min": null, "range_max": null, "nominal_min": null, "nominal_max": null, "zero_point": null, "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "continuous_range": null }, "group_rank": null, "description": "In October 2022, the US instituted [new, wide-ranging export controls](https://www.csis.org/analysis/choking-chinas-access-future-ai) aimed at controlling a set of “chokepoint” technologies in the global semiconductor supply chain. China is one of the leading countries in AI, but AI advancements have historically run on compute, and China’s access to cost-competitive machine learning (ML) compute — the Chinese AI industry currently relies in large part on AI chips designed and fabricated abroad, and in particular Nvidia GPUs — is now in question.\n\nThe [US export controls](https://www.ecfr.gov/current/title-15/subtitle-B/chapter-VII/subchapter-C) prevent the export of high-performance AI chips to China, as well as the export of advanced tooling and materials used to make such chips. Still, one way for China to gain better access to cost-competitive ML compute is to indigenously produce high-quality AI chips. In 2015, China set out an ambitious plan aiming to bring the proportion of chips bought by Chinese firms that is domestically produced from 15% to 70% by 2025; as of 2021 [that number had gone up to 24%](https://asia.nikkei.com/Politics/International-relations/US-China-tensions/U.S.-tech-curbs-threaten-China-s-quest-for-chip-independence). With the October restrictions — and with US allies like Taiwan, Japan, South Korea, and the Netherlands enacting their own controls — Chinese firms now face substantial challenges in their attempts to indigenize semiconductor production.\n\nChips are fabricated in “fabs” — semiconductor fabrication plants where the chips are manufactured on silicon wafers. Fabs are generally owned and operated by “(pure-play) foundries” — companies that specialize in chip fabrication for clients (who design the chips) on a contractual basis. (The alternative to the foundry model is an integrated device manufacturer (ISM) which both designs and fabricates chips. Examples of ISMs include Intel and SK Hynix.)\n\nThe most famous, advanced, and profitable foundry is [Taiwan Semiconductor Manufacturing Company (TSMC)](https://www.google.com/finance/quote/2330:TPE) which in Q1 2023 had a [market share](https://www.trendforce.com/presscenter/news/20230612-11719.html) of 60.1%. As of May 2023, TSMC has a market cap of 15T TWD ($470B). Notable Chinese foundries include [Semiconductor Manufacturing International Corporation (SMIC)](https://www.google.com/finance/quote/688981:SHA) (Q1 2023 market share: 5.3%) and [Hua Hong Semiconductor Limited](https://www.google.com/finance/quote/1347:HKG) (Q1 2023 market share: 3.0%). Due to the October export controls, SMIC and Hua Hong are cut off from much of the cutting-edge equipment and materials necessary to fabricate advanced chips.\n\nFor more information, I have written a primer on [indigenously made Chinese AI chips](https://docs.google.com/document/d/1r-aLT67d2OQnuhZJZihkr0KxxB-PsDMKMut2flaYQ2c/edit?usp=sharing), intended to provide a useful introduction and reference for anyone interested in forecasting or learning more about China’s ability to indigenously produce AI chips. It explains some key concepts, provides an overview of the chip-making process, provides an overview of relevant export controls, and describes key inputs, context, and organizations relevant to understanding Chinese chip-making progress.", "resolution_criteria": "The question will resolve as **Yes** if any Chinese semiconductor foundry has 20% or more semiconductor foundry revenue market share in any one quarter before January 1st, 2033. The question will resolve as **No** otherwise.", "fine_print": "A “Chinese semiconductor foundry” is a firm that (a) primarily fabricates semiconductors for customers, and (b) is headquartered in China. (It can be a subsidiary of another firm so long as those conditions are met.) The most prominent examples currently are SMIC and Hua Hong.\n\nIf the PRC annexes Taiwan, then annexed-Taiwan, including the chips being produced there, counts as being part of China.\n\nThe question will be resolved using data from Statista or, if that isn’t available, from another credible source.", "post_id": 20157, "aggregations": { "recency_weighted": { "history": [ { "start_time": 1758502497.586691, "end_time": 1759143175.561826, "forecaster_count": 52, "interval_lower_bounds": [ 0.55 ], "centers": [ 0.7 ], "interval_upper_bounds": [ 0.77 ] } ], "latest": { "start_time": 1758502497.586691, "end_time": 1759143175.561826, "forecaster_count": 52, "interval_lower_bounds": [ 0.55 ], "centers": [ 0.7 ], "interval_upper_bounds": [ 0.77 ], "forecast_values": [ 0.30000000000000004, 0.7 ], "means": [ 0.6385470879890909 ], "histogram": [ [ 0.0, 0.1333270524573957, 0.0, 0.0, 0.0, 0.01040622348426393, 0.41207610303291553, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.4817346416537838, 0.1138807454632285, 0.0, 0.0, 0.0, 0.0, 0.027172461172235558, 0.004173285167582051, 0.2515406540753948, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 1.0505334438371767, 0.0, 0.0, 0.09784722664542772, 0.2002418601215102, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06463710864254305, 0.0, 0.0, 0.0, 0.05771699504164168, 0.0, 0.0, 0.0, 0.0, 0.0, 0.3638820728450715, 0.031134300295554664, 0.0, 0.38056261610345776, 0.0, 0.2974800112984262, 0.08934312118698887, 0.6513080407565454, 0.14666562708809194, 0.0, 0.0, 0.17659841324481915, 1.3653311852731553, 0.0, 0.04559318956273326, 1.1295374919159165, 0.0, 0.23072242730151374, 0.0, 0.8693280352332947, 0.9515091831531896, 0.0, 0.7535956555594016, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.3723102402239968, 0.0, 0.012491892736397447, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.002007023398223986, 0.0, 2.154214787738731 ] ] }, "score_data": {}, "movement": null }, "unweighted": { "history": [], "latest": null, "score_data": {}, "movement": null }, "single_aggregation": { "history": [], "latest": null, "score_data": {}, "movement": null }, "metaculus_prediction": { "history": [ { "start_time": 1728287850.110762, "end_time": null, "forecaster_count": 45, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null } ], "latest": { "start_time": 1728287850.110762, "end_time": null, "forecaster_count": 45, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null, "forecast_values": [ 0.6223981485356453, 0.3776018514643546 ], "means": null, "histogram": null }, "score_data": {}, "movement": null } } }, "user_permission": "forecaster", "vote": { "score": 12, "user_vote": null }, "forecasts_count": 114, "key_factors": [], "is_current_content_translated": false, "description": "In October 2022, the US instituted [new, wide-ranging export controls](https://www.csis.org/analysis/choking-chinas-access-future-ai) aimed at controlling a set of “chokepoint” technologies in the global semiconductor supply chain. China is one of the leading countries in AI, but AI advancements have historically run on compute, and China’s access to cost-competitive machine learning (ML) compute — the Chinese AI industry currently relies in large part on AI chips designed and fabricated abroad, and in particular Nvidia GPUs — is now in question.\n\nThe [US export controls](https://www.ecfr.gov/current/title-15/subtitle-B/chapter-VII/subchapter-C) prevent the export of high-performance AI chips to China, as well as the export of advanced tooling and materials used to make such chips. Still, one way for China to gain better access to cost-competitive ML compute is to indigenously produce high-quality AI chips. In 2015, China set out an ambitious plan aiming to bring the proportion of chips bought by Chinese firms that is domestically produced from 15% to 70% by 2025; as of 2021 [that number had gone up to 24%](https://asia.nikkei.com/Politics/International-relations/US-China-tensions/U.S.-tech-curbs-threaten-China-s-quest-for-chip-independence). With the October restrictions — and with US allies like Taiwan, Japan, South Korea, and the Netherlands enacting their own controls — Chinese firms now face substantial challenges in their attempts to indigenize semiconductor production.\n\nChips are fabricated in “fabs” — semiconductor fabrication plants where the chips are manufactured on silicon wafers. Fabs are generally owned and operated by “(pure-play) foundries” — companies that specialize in chip fabrication for clients (who design the chips) on a contractual basis. (The alternative to the foundry model is an integrated device manufacturer (ISM) which both designs and fabricates chips. Examples of ISMs include Intel and SK Hynix.)\n\nThe most famous, advanced, and profitable foundry is [Taiwan Semiconductor Manufacturing Company (TSMC)](https://www.google.com/finance/quote/2330:TPE) which in Q1 2023 had a [market share](https://www.trendforce.com/presscenter/news/20230612-11719.html) of 60.1%. As of May 2023, TSMC has a market cap of 15T TWD ($470B). Notable Chinese foundries include [Semiconductor Manufacturing International Corporation (SMIC)](https://www.google.com/finance/quote/688981:SHA) (Q1 2023 market share: 5.3%) and [Hua Hong Semiconductor Limited](https://www.google.com/finance/quote/1347:HKG) (Q1 2023 market share: 3.0%). Due to the October export controls, SMIC and Hua Hong are cut off from much of the cutting-edge equipment and materials necessary to fabricate advanced chips.\n\nFor more information, I have written a primer on [indigenously made Chinese AI chips](https://docs.google.com/document/d/1r-aLT67d2OQnuhZJZihkr0KxxB-PsDMKMut2flaYQ2c/edit?usp=sharing), intended to provide a useful introduction and reference for anyone interested in forecasting or learning more about China’s ability to indigenously produce AI chips. It explains some key concepts, provides an overview of the chip-making process, provides an overview of relevant export controls, and describes key inputs, context, and organizations relevant to understanding Chinese chip-making progress." }, { "id": 20146, "title": "Will a Chinese fab have a volume production process that uses EUV photolithography before 2033?", "short_title": "Chinese volume production using EUV by 2033?", "url_title": "Chinese volume production using EUV by 2033?", "slug": "chinese-volume-production-using-euv-by-2033", "author_id": 111541, "author_username": "DerMannOhneEigenschaften", "coauthors": [], "created_at": "2023-11-24T13:08:16.005256Z", "published_at": "2023-12-06T11:00:00Z", "edited_at": "2025-09-10T21:56:22.877862Z", "curation_status": "approved", "curation_status_updated_at": "2023-12-06T11:00:00Z", "comment_count": 0, "status": "open", "resolved": false, "actual_close_time": null, "scheduled_close_time": "2033-01-01T11:00:00Z", "scheduled_resolve_time": "2033-01-01T11:00:00Z", "actual_resolve_time": null, "open_time": "2023-12-06T11:00:00Z", "nr_forecasters": 35, "html_metadata_json": null, "projects": { "site_main": [ { "id": 144, "type": "site_main", "name": "Metaculus Community", "slug": null, "header_image": null, "prize_pool": null, "start_date": null, "close_date": null, "forecasting_end_date": null, "html_metadata_json": null, "is_ongoing": null, "user_permission": "forecaster", "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2025-07-18T17:28:18.838588Z", "score_type": null, "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } ], "tournament": [ { "id": 2784, "type": "tournament", "name": "Chinese AI Chips", "slug": "chinese-ai-chips", "header_image": "https://metaculus-web-media.s3.amazonaws.com/cac-new_1.webp", "prize_pool": null, "start_date": "2023-12-06T12:00:33Z", "close_date": "2033-01-02T23:59:33Z", "forecasting_end_date": "2033-01-01T11:00:00Z", "html_metadata_json": null, "is_ongoing": true, "user_permission": null, "created_at": "2023-11-28T16:30:32.190353Z", "edited_at": "2025-09-22T00:55:08.932036Z", "score_type": "peer_tournament", "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } ], "default_project": { "id": 2784, "type": "tournament", "name": "Chinese AI Chips", "slug": "chinese-ai-chips", "header_image": "https://metaculus-web-media.s3.amazonaws.com/cac-new_1.webp", "prize_pool": null, "start_date": "2023-12-06T12:00:33Z", "close_date": "2033-01-02T23:59:33Z", "forecasting_end_date": "2033-01-01T11:00:00Z", "html_metadata_json": null, "is_ongoing": true, "user_permission": null, "created_at": "2023-11-28T16:30:32.190353Z", "edited_at": "2025-09-22T00:55:08.932036Z", "score_type": "peer_tournament", "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" }, "category": [ { "id": 3701, "name": "Technology", "slug": "technology", "emoji": "⚙️", "description": "Technology", "type": "category" }, { "id": 3692, "name": "Computing and Math", "slug": "computing-and-math", "emoji": "💻", "description": "Computing and Math", "type": "category" } ], "question_series": [ { "id": 2343, "type": "question_series", "name": "AI in China", "slug": "ai-in-china", "header_image": "https://metaculus-web-media.s3.amazonaws.com/ai-china.png", "prize_pool": null, "start_date": null, "close_date": null, "forecasting_end_date": null, "html_metadata_json": null, "is_ongoing": true, "user_permission": "forecaster", "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2025-09-22T00:55:09.020170Z", "score_type": "relative_legacy_tournament", "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } ] }, "question": { "id": 20146, "title": "Will a Chinese fab have a volume production process that uses EUV photolithography before 2033?", "created_at": "2023-11-24T13:08:16.005256Z", "open_time": "2023-12-06T11:00:00Z", "cp_reveal_time": "2023-12-20T11:00:00Z", "spot_scoring_time": "2023-12-20T11:00:00Z", "scheduled_resolve_time": "2033-01-01T11:00:00Z", "actual_resolve_time": null, "resolution_set_time": null, "scheduled_close_time": "2033-01-01T11:00:00Z", "actual_close_time": "2033-01-01T11:00:00Z", "type": "binary", "options": null, "group_variable": "", "status": "open", "possibilities": { "type": "binary" }, "resolution": null, "include_bots_in_aggregates": false, "question_weight": 1.0, "default_score_type": "peer", "default_aggregation_method": "recency_weighted", "label": "", "unit": "", "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "scaling": { "range_min": null, "range_max": null, "nominal_min": null, "nominal_max": null, "zero_point": null, "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "continuous_range": null }, "group_rank": null, "description": "In October 2022, the US instituted [new, wide-ranging export controls](https://www.csis.org/analysis/choking-chinas-access-future-ai) aimed at controlling a set of “chokepoint” technologies in the global semiconductor supply chain. China is one of the leading countries in AI, but AI advancements have historically run on compute, and China’s access to cost-competitive machine learning (ML) compute — the Chinese AI industry currently relies in large part on AI chips designed and fabricated abroad, and in particular Nvidia GPUs — is now in question.\n\nThe [US export controls](https://www.ecfr.gov/current/title-15/subtitle-B/chapter-VII/subchapter-C) prevent the export of high-performance AI chips to China, as well as the export of advanced tooling and materials used to make such chips. Still, one way for China to gain better access to cost-competitive ML compute is to indigenously produce high-quality AI chips. In 2015, China set out an ambitious plan aiming to bring the proportion of chips bought by Chinese firms that is domestically produced from 15% to 70% by 2025; as of 2021 [that number had gone up to 24%](https://asia.nikkei.com/Politics/International-relations/US-China-tensions/U.S.-tech-curbs-threaten-China-s-quest-for-chip-independence). With the October restrictions — and with US allies like Taiwan, Japan, South Korea, and the Netherlands enacting their own controls — Chinese firms now face substantial challenges in their attempts to indigenize semiconductor production.\n\nChips are fabricated in “fabs” — semiconductor fabrication plants where the chips are manufactured on silicon wafers. Notable Chinese semiconductor manufacturing companies include [Semiconductor Manufacturing International Corporation (SMIC)](https://finance.yahoo.com/quote/0981.HK?p=0981.HK&.tsrc=fin-srch) and [Hua Hong Semiconductor Limited](https://finance.yahoo.com/quote/1347.HK?p=1347.HK&.tsrc=fin-srch).\n\nCritically, SMIC and Hua Hong rely on foreign-made [photolithography](https://en.wikipedia.org/wiki/Photolithography) machines, in particular deep ultraviolet (DUV) immersion machines made by ASML (the Netherlands) and Nikon (Japan). These devices are some of the most complex machinery ever made by humans. The most advanced type of photolithography machine – extreme ultraviolet (EUV) – is produced only by ASML, and took three decades to develop. As [Miller](https://www.christophermiller.net/semiconductors-1) writes, “EUV machines have multiple components that, on their own, constitute epically complex engineering challenges. Replicating just the laser in an EUV system requires perfectly identifying and assembling 457,329 parts. A single defect could cause debilitating delays or reliability problems.” Dutch export controls prohibit ASML from exporting EUV (and some DUV) lithography machines to China.\n\n(This question concerns volume production. Usually, a company will begin by producing small volumes of chips on a new process, called \"risk production\", and only later scale up once the process has been refined. For reference, [it took TSMC](https://www.semianalysis.com/p/semiconductor-fab-buildout-delays) about a year to reach 45,000 wafers per month after the initial shipment from its N3 process node, and about six months for each of its N5 and N7 processes to do the same.)\n\nChina’s major photolithography maker is Shanghai Micro Electronics Equipment (SMEE), founded in the early 2000s. SMEE [markets](http://www.smee.com.cn/eis.pub?service=homepageService&method=indexinfo&onclicknodeno=1_4_4_1) krypton fluoride (KrF) and argon fluoride (ArF) dry steppers for mature process nodes, but these either aren’t ready for mass production and/or are far behind (>15 years) Dutch and Japanese KrF and ArF machines in terms of resolution, throughput, and quality. SMEE does not market any more advanced argon fluoride immersion (ArFi) machines, let alone any EUV machines.\n\nFor more information, I have written a primer on [indigenously made Chinese AI chips](https://docs.google.com/document/d/1r-aLT67d2OQnuhZJZihkr0KxxB-PsDMKMut2flaYQ2c/edit?usp=sharing), intended to provide a useful introduction and reference for anyone interested in forecasting or learning more about China’s ability to indigenously produce AI chips. It explains some key concepts, provides an overview of the chip-making process, provides an overview of relevant export controls, and describes key inputs, context, and organizations relevant to understanding Chinese chip-making progress.", "resolution_criteria": "The question will resolve as **Yes** if a Chinese fab has achieved volume production using extreme ultraviolet (EUV) photolithography before January 1st, 2033. The question will resolve as **No** otherwise. If the PRC annexes Taiwan, then annexed-Taiwan counts as being part of China (i.e., fabs in annexed-Taiwan are considered Chinese fabs).\n\nThe question resolves as **Yes** if and only if credible sources report that a Chinese fab has achieved volume production using EUV photolithography, and/or if the fab itself claims it produces 50,000 or more wafers per month using EUV photolithography. The news reports must mention that the fab has achieved “(high) volume production” or “mass production”.", "fine_print": "A “Chinese fab” is a semiconductor fabrication facility located in China and owned by a company that is headquartered in China. “China” includes mainland China, Hong Kong, and Macao (but not Taiwan). If the PRC annexes Taiwan, then annexed-Taiwan counts as being part of China (i.e., fabs in annexed-Taiwan are considered Chinese fabs).", "post_id": 20146, "aggregations": { "recency_weighted": { "history": [ { "start_time": 1757541372.641761, "end_time": 1759164227.956336, "forecaster_count": 33, "interval_lower_bounds": [ 0.75 ], "centers": [ 0.9 ], "interval_upper_bounds": [ 0.95 ] } ], "latest": { "start_time": 1757541372.641761, "end_time": 1759164227.956336, "forecaster_count": 33, "interval_lower_bounds": [ 0.75 ], "centers": [ 0.9 ], "interval_upper_bounds": [ 0.95 ], "forecast_values": [ 0.09999999999999998, 0.9 ], "means": [ 0.8019014358813159 ], "histogram": [ [ 0.0, 0.0, 0.0, 0.0, 0.31286391742790765, 0.07560105912010007, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.5778677609141162, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.018087904224968542, 0.0, 0.0, 0.0, 0.0, 0.06427640630688772, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.09869070413164345, 0.0, 0.1538804463707397, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1349426696819013, 0.0, 0.0, 1.0541425161069093, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.34848960163546905, 0.0, 0.47494197203925326, 0.0, 0.0, 0.3683695207813824, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.429306930151368, 0.0, 0.0, 0.9968156189350071, 0.05022829445599939, 0.8379487954001852, 0.0, 1.2224742237437058, 0.38723206162456447, 0.0, 0.0, 0.0, 2.394205877294843 ] ] }, "score_data": {}, "movement": null }, "unweighted": { "history": [], "latest": null, "score_data": {}, "movement": null }, "single_aggregation": { "history": [], "latest": null, "score_data": {}, "movement": null }, "metaculus_prediction": { "history": [ { "start_time": 1728287083.615452, "end_time": null, "forecaster_count": 26, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null } ], "latest": { "start_time": 1728287083.615452, "end_time": null, "forecaster_count": 26, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null, "forecast_values": [ 0.21461347772413664, 0.7853865222758634 ], "means": null, "histogram": null }, "score_data": {}, "movement": null } } }, "user_permission": "forecaster", "vote": { "score": 11, "user_vote": null }, "forecasts_count": 100, "key_factors": [], "is_current_content_translated": false, "description": "In October 2022, the US instituted [new, wide-ranging export controls](https://www.csis.org/analysis/choking-chinas-access-future-ai) aimed at controlling a set of “chokepoint” technologies in the global semiconductor supply chain. China is one of the leading countries in AI, but AI advancements have historically run on compute, and China’s access to cost-competitive machine learning (ML) compute — the Chinese AI industry currently relies in large part on AI chips designed and fabricated abroad, and in particular Nvidia GPUs — is now in question.\n\nThe [US export controls](https://www.ecfr.gov/current/title-15/subtitle-B/chapter-VII/subchapter-C) prevent the export of high-performance AI chips to China, as well as the export of advanced tooling and materials used to make such chips. Still, one way for China to gain better access to cost-competitive ML compute is to indigenously produce high-quality AI chips. In 2015, China set out an ambitious plan aiming to bring the proportion of chips bought by Chinese firms that is domestically produced from 15% to 70% by 2025; as of 2021 [that number had gone up to 24%](https://asia.nikkei.com/Politics/International-relations/US-China-tensions/U.S.-tech-curbs-threaten-China-s-quest-for-chip-independence). With the October restrictions — and with US allies like Taiwan, Japan, South Korea, and the Netherlands enacting their own controls — Chinese firms now face substantial challenges in their attempts to indigenize semiconductor production.\n\nChips are fabricated in “fabs” — semiconductor fabrication plants where the chips are manufactured on silicon wafers. Notable Chinese semiconductor manufacturing companies include [Semiconductor Manufacturing International Corporation (SMIC)](https://finance.yahoo.com/quote/0981.HK?p=0981.HK&.tsrc=fin-srch) and [Hua Hong Semiconductor Limited](https://finance.yahoo.com/quote/1347.HK?p=1347.HK&.tsrc=fin-srch).\n\nCritically, SMIC and Hua Hong rely on foreign-made [photolithography](https://en.wikipedia.org/wiki/Photolithography) machines, in particular deep ultraviolet (DUV) immersion machines made by ASML (the Netherlands) and Nikon (Japan). These devices are some of the most complex machinery ever made by humans. The most advanced type of photolithography machine – extreme ultraviolet (EUV) – is produced only by ASML, and took three decades to develop. As [Miller](https://www.christophermiller.net/semiconductors-1) writes, “EUV machines have multiple components that, on their own, constitute epically complex engineering challenges. Replicating just the laser in an EUV system requires perfectly identifying and assembling 457,329 parts. A single defect could cause debilitating delays or reliability problems.” Dutch export controls prohibit ASML from exporting EUV (and some DUV) lithography machines to China.\n\n(This question concerns volume production. Usually, a company will begin by producing small volumes of chips on a new process, called \"risk production\", and only later scale up once the process has been refined. For reference, [it took TSMC](https://www.semianalysis.com/p/semiconductor-fab-buildout-delays) about a year to reach 45,000 wafers per month after the initial shipment from its N3 process node, and about six months for each of its N5 and N7 processes to do the same.)\n\nChina’s major photolithography maker is Shanghai Micro Electronics Equipment (SMEE), founded in the early 2000s. SMEE [markets](http://www.smee.com.cn/eis.pub?service=homepageService&method=indexinfo&onclicknodeno=1_4_4_1) krypton fluoride (KrF) and argon fluoride (ArF) dry steppers for mature process nodes, but these either aren’t ready for mass production and/or are far behind (>15 years) Dutch and Japanese KrF and ArF machines in terms of resolution, throughput, and quality. SMEE does not market any more advanced argon fluoride immersion (ArFi) machines, let alone any EUV machines.\n\nFor more information, I have written a primer on [indigenously made Chinese AI chips](https://docs.google.com/document/d/1r-aLT67d2OQnuhZJZihkr0KxxB-PsDMKMut2flaYQ2c/edit?usp=sharing), intended to provide a useful introduction and reference for anyone interested in forecasting or learning more about China’s ability to indigenously produce AI chips. It explains some key concepts, provides an overview of the chip-making process, provides an overview of relevant export controls, and describes key inputs, context, and organizations relevant to understanding Chinese chip-making progress." }, { "id": 20136, "title": "Will any Chinese semiconductor foundry have at least 20% of the market cap that TSMC has in 2028?", "short_title": "Chinese foundry with ≥0.2x of TSMC market cap", "url_title": "Chinese foundry with ≥0.2x of TSMC market cap", "slug": "chinese-foundry-with-02x-of-tsmc-market-cap", "author_id": 111541, "author_username": "DerMannOhneEigenschaften", "coauthors": [], "created_at": "2023-11-24T10:48:16.728916Z", "published_at": "2023-12-06T11:00:00Z", "edited_at": "2025-09-10T10:34:32.273065Z", "curation_status": "approved", "curation_status_updated_at": "2023-12-06T11:00:00Z", "comment_count": 2, "status": "open", "resolved": false, "actual_close_time": null, "scheduled_close_time": "2028-01-01T11:00:00Z", "scheduled_resolve_time": "2028-01-01T11:00:00Z", "actual_resolve_time": null, "open_time": "2023-12-06T11:00:00Z", "nr_forecasters": 46, "html_metadata_json": null, "projects": { "site_main": [ { "id": 144, "type": "site_main", "name": "Metaculus Community", "slug": null, "header_image": null, "prize_pool": null, "start_date": null, "close_date": null, "forecasting_end_date": null, "html_metadata_json": null, "is_ongoing": null, "user_permission": "forecaster", "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2025-07-18T17:28:18.838588Z", "score_type": null, "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } ], "tournament": [ { "id": 2784, "type": "tournament", "name": "Chinese AI Chips", "slug": "chinese-ai-chips", "header_image": "https://metaculus-web-media.s3.amazonaws.com/cac-new_1.webp", "prize_pool": null, "start_date": "2023-12-06T12:00:33Z", "close_date": "2033-01-02T23:59:33Z", "forecasting_end_date": "2033-01-01T11:00:00Z", "html_metadata_json": null, "is_ongoing": true, "user_permission": null, "created_at": "2023-11-28T16:30:32.190353Z", "edited_at": "2025-09-22T00:55:08.932036Z", "score_type": "peer_tournament", "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } ], "default_project": { "id": 2784, "type": "tournament", "name": "Chinese AI Chips", "slug": "chinese-ai-chips", "header_image": "https://metaculus-web-media.s3.amazonaws.com/cac-new_1.webp", "prize_pool": null, "start_date": "2023-12-06T12:00:33Z", "close_date": "2033-01-02T23:59:33Z", "forecasting_end_date": "2033-01-01T11:00:00Z", "html_metadata_json": null, "is_ongoing": true, "user_permission": null, "created_at": "2023-11-28T16:30:32.190353Z", "edited_at": "2025-09-22T00:55:08.932036Z", "score_type": "peer_tournament", "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" }, "category": [ { "id": 3701, "name": "Technology", "slug": "technology", "emoji": "⚙️", "description": "Technology", "type": "category" }, { "id": 3692, "name": "Computing and Math", "slug": "computing-and-math", "emoji": "💻", "description": "Computing and Math", "type": "category" } ], "question_series": [ { "id": 2343, "type": "question_series", "name": "AI in China", "slug": "ai-in-china", "header_image": "https://metaculus-web-media.s3.amazonaws.com/ai-china.png", "prize_pool": null, "start_date": null, "close_date": null, "forecasting_end_date": null, "html_metadata_json": null, "is_ongoing": true, "user_permission": "forecaster", "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2025-09-22T00:55:09.020170Z", "score_type": "relative_legacy_tournament", "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } ] }, "question": { "id": 20136, "title": "Will any Chinese semiconductor foundry have at least 20% of the market cap that TSMC has in 2028?", "created_at": "2023-11-24T10:48:16.728916Z", "open_time": "2023-12-06T11:00:00Z", "cp_reveal_time": "2023-12-20T11:00:00Z", "spot_scoring_time": "2023-12-20T11:00:00Z", "scheduled_resolve_time": "2028-01-01T11:00:00Z", "actual_resolve_time": null, "resolution_set_time": null, "scheduled_close_time": "2028-01-01T11:00:00Z", "actual_close_time": "2028-01-01T11:00:00Z", "type": "binary", "options": null, "group_variable": "", "status": "open", "possibilities": { "type": "binary" }, "resolution": null, "include_bots_in_aggregates": false, "question_weight": 1.0, "default_score_type": "peer", "default_aggregation_method": "recency_weighted", "label": "", "unit": "", "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "scaling": { "range_min": null, "range_max": null, "nominal_min": null, "nominal_max": null, "zero_point": null, "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "continuous_range": null }, "group_rank": null, "description": "In October 2022, the US instituted [new, wide-ranging export controls](https://www.csis.org/analysis/choking-chinas-access-future-ai) aimed at controlling a set of “chokepoint” technologies in the global semiconductor supply chain. China is one of the leading countries in AI, but AI advancements have historically run on compute, and China’s access to cost-competitive machine learning (ML) compute — the Chinese AI industry currently relies in large part on AI chips designed and fabricated abroad, and in particular Nvidia GPUs — is now in question.\n\nThe [US export controls](https://www.ecfr.gov/current/title-15/subtitle-B/chapter-VII/subchapter-C) prevent the export of high-performance AI chips to China, as well as the export of advanced tooling and materials used to make such chips. Still, one way for China to gain better access to cost-competitive ML compute is to indigenously produce high-quality AI chips. In 2015, China set out an ambitious plan aiming to bring the proportion of chips bought by Chinese firms that is domestically produced from 15% to 70% by 2025; as of 2021 [that number had gone up to 24%](https://asia.nikkei.com/Politics/International-relations/US-China-tensions/U.S.-tech-curbs-threaten-China-s-quest-for-chip-independence). With the October restrictions — and with US allies like Taiwan, Japan, South Korea, and the Netherlands enacting their own controls — Chinese firms now face substantial challenges in their attempts to indigenize semiconductor production.\n\nChips are fabricated in “fabs” — semiconductor fabrication plants where the chips are manufactured on silicon wafers. Fabs are generally owned and operated by “(pure-play) foundries” — companies that specialize in chip fabrication for clients (who design the chips) on a contractual basis. (The alternative to the foundry model is an integrated device manufacturer (ISM) which both designs and fabricates chips. Examples of ISMs include Intel and SK Hynix.)\n\nThe most famous, advanced, and profitable foundry is [Taiwan Semiconductor Manufacturing Company (TSMC)](https://www.google.com/finance/quote/2330:TPE) As of May 2023, TSMC has a market cap of 15T TWD ($470B). Notable Chinese foundries include [Semiconductor Manufacturing International Corporation (SMIC)](https://www.google.com/finance/quote/688981:SHA) and [Hua Hong Semiconductor Limited](https://www.google.com/finance/quote/1347:HKG), which have market caps of 220B CNY ($31B) and 43B HKD ($5.5B) respectively, meaning 7% and 1% of TSMC.\n\nFor more information, I have written a primer on [indigenously made Chinese AI chips](https://docs.google.com/document/d/1r-aLT67d2OQnuhZJZihkr0KxxB-PsDMKMut2flaYQ2c/edit?usp=sharing), intended to provide a useful introduction and reference for anyone interested in forecasting or learning more about China’s ability to indigenously produce AI chips. It explains some key concepts, provides an overview of the chip-making process, provides an overview of relevant export controls, and describes key inputs, context, and organizations relevant to understanding Chinese chip-making progress.", "resolution_criteria": "The question will resolve as **Yes** if any Chinese semiconductor foundry has a market cap (measured in US dollars) that is at least 20% as large as the market capitalization of Taiwan Semiconductor Manufacturing Company (TSMC), on January 1st, 2028. The question will resolve as **No** if no Chinese foundry has a market capitalization that is 20% ore more of TSMC’s market cap, on that same date. If TSMC no longer exists, or if it is no longer publicly traded, the question will be annulled.", "fine_print": "A “Chinese semiconductor foundry” is a firm that (a) primarily fabricates semiconductors for customers, and (b) is headquartered in China. (It can be a subsidiary of another firm so long as those conditions are met.) The most prominent examples currently are SMIC and Hua Hong.\n\nIf the PRC annexes Taiwan, then annexed-Taiwan counts as being part of China (i.e., foundries in annexed-Taiwan — including TSMC, if it still exists — are considered Chinese foundries).\n\nTo determine each firm’s “market capitalization”, we use figures reported by [Google Finance](https://www.google.com/finance/), or, if those are not available, by another reputable source. If a firm is listed on multiple exchanges, we use whatever is the firm’s maximum market capitalization on any exchange.", "post_id": 20136, "aggregations": { "recency_weighted": { "history": [ { "start_time": 1758141560.163537, "end_time": 1759334515.710306, "forecaster_count": 23, "interval_lower_bounds": [ 0.3 ], "centers": [ 0.33 ], "interval_upper_bounds": [ 0.68 ] } ], "latest": { "start_time": 1758141560.163537, "end_time": 1759334515.710306, "forecaster_count": 23, "interval_lower_bounds": [ 0.3 ], "centers": [ 0.33 ], "interval_upper_bounds": [ 0.68 ], "forecast_values": [ 0.6699999999999999, 0.33 ], "means": [ 0.5018729415633161 ], "histogram": [ [ 0.0, 0.0, 0.0, 0.13981930080831617, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.22781833750170227, 0.0, 0.4512058906735414, 0.0, 0.0, 0.30413601865739065, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.26402013543029734, 0.0, 0.0, 0.0, 0.0, 0.0, 1.3010537000243148, 0.0, 0.0, 1.5103156119575218, 0.0, 0.0, 0.1659893709142453, 0.0, 0.04671076085250769, 0.0, 0.19523450305981643, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1164748146920504, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.5751117920109702, 0.0, 0.7234704561894825, 0.0, 0.3484801054844811, 0.0, 0.0, 0.0, 0.0, 0.0, 0.061064077012331824, 0.0339924118233929, 0.0, 0.0, 0.07732302163410486, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 1.568429481272093 ] ] }, "score_data": {}, "movement": null }, "unweighted": { "history": [], "latest": null, "score_data": {}, "movement": null }, "single_aggregation": { "history": [], "latest": null, "score_data": {}, "movement": null }, "metaculus_prediction": { "history": [ { "start_time": 1728288762.09922, "end_time": null, "forecaster_count": 36, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null } ], "latest": { "start_time": 1728288762.09922, "end_time": null, "forecaster_count": 36, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null, "forecast_values": [ 0.859013224164868, 0.140986775835132 ], "means": null, "histogram": null }, "score_data": {}, "movement": null } } }, "user_permission": "forecaster", "vote": { "score": 12, "user_vote": null }, "forecasts_count": 146, "key_factors": [], "is_current_content_translated": false, "description": "In October 2022, the US instituted [new, wide-ranging export controls](https://www.csis.org/analysis/choking-chinas-access-future-ai) aimed at controlling a set of “chokepoint” technologies in the global semiconductor supply chain. China is one of the leading countries in AI, but AI advancements have historically run on compute, and China’s access to cost-competitive machine learning (ML) compute — the Chinese AI industry currently relies in large part on AI chips designed and fabricated abroad, and in particular Nvidia GPUs — is now in question.\n\nThe [US export controls](https://www.ecfr.gov/current/title-15/subtitle-B/chapter-VII/subchapter-C) prevent the export of high-performance AI chips to China, as well as the export of advanced tooling and materials used to make such chips. Still, one way for China to gain better access to cost-competitive ML compute is to indigenously produce high-quality AI chips. In 2015, China set out an ambitious plan aiming to bring the proportion of chips bought by Chinese firms that is domestically produced from 15% to 70% by 2025; as of 2021 [that number had gone up to 24%](https://asia.nikkei.com/Politics/International-relations/US-China-tensions/U.S.-tech-curbs-threaten-China-s-quest-for-chip-independence). With the October restrictions — and with US allies like Taiwan, Japan, South Korea, and the Netherlands enacting their own controls — Chinese firms now face substantial challenges in their attempts to indigenize semiconductor production.\n\nChips are fabricated in “fabs” — semiconductor fabrication plants where the chips are manufactured on silicon wafers. Fabs are generally owned and operated by “(pure-play) foundries” — companies that specialize in chip fabrication for clients (who design the chips) on a contractual basis. (The alternative to the foundry model is an integrated device manufacturer (ISM) which both designs and fabricates chips. Examples of ISMs include Intel and SK Hynix.)\n\nThe most famous, advanced, and profitable foundry is [Taiwan Semiconductor Manufacturing Company (TSMC)](https://www.google.com/finance/quote/2330:TPE) As of May 2023, TSMC has a market cap of 15T TWD ($470B). Notable Chinese foundries include [Semiconductor Manufacturing International Corporation (SMIC)](https://www.google.com/finance/quote/688981:SHA) and [Hua Hong Semiconductor Limited](https://www.google.com/finance/quote/1347:HKG), which have market caps of 220B CNY ($31B) and 43B HKD ($5.5B) respectively, meaning 7% and 1% of TSMC.\n\nFor more information, I have written a primer on [indigenously made Chinese AI chips](https://docs.google.com/document/d/1r-aLT67d2OQnuhZJZihkr0KxxB-PsDMKMut2flaYQ2c/edit?usp=sharing), intended to provide a useful introduction and reference for anyone interested in forecasting or learning more about China’s ability to indigenously produce AI chips. It explains some key concepts, provides an overview of the chip-making process, provides an overview of relevant export controls, and describes key inputs, context, and organizations relevant to understanding Chinese chip-making progress." }, { "id": 20135, "title": "Will a Chinese firm market an EUV photolithography machine before 2033?", "short_title": "Chinese EUV machine marketed by 2033?", "url_title": "Chinese EUV machine marketed by 2033?", "slug": "chinese-euv-machine-marketed-by-2033", "author_id": 111541, "author_username": "DerMannOhneEigenschaften", "coauthors": [], "created_at": "2023-11-24T10:38:48.175836Z", "published_at": "2023-12-06T11:00:00Z", "edited_at": "2025-09-17T15:01:28.955194Z", "curation_status": "approved", "curation_status_updated_at": "2023-12-06T11:00:00Z", "comment_count": 0, "status": "open", "resolved": false, "actual_close_time": null, "scheduled_close_time": "2033-01-01T11:00:00Z", "scheduled_resolve_time": "2033-01-01T11:00:00Z", "actual_resolve_time": null, "open_time": "2023-12-06T11:00:00Z", "nr_forecasters": 38, "html_metadata_json": null, "projects": { "site_main": [ { "id": 144, "type": "site_main", "name": "Metaculus Community", "slug": null, "header_image": null, "prize_pool": null, "start_date": null, "close_date": null, "forecasting_end_date": null, "html_metadata_json": null, "is_ongoing": null, "user_permission": "forecaster", "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2025-07-18T17:28:18.838588Z", "score_type": null, "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } ], "tournament": [ { "id": 2784, "type": "tournament", "name": "Chinese AI Chips", "slug": "chinese-ai-chips", "header_image": "https://metaculus-web-media.s3.amazonaws.com/cac-new_1.webp", "prize_pool": null, "start_date": "2023-12-06T12:00:33Z", "close_date": "2033-01-02T23:59:33Z", "forecasting_end_date": "2033-01-01T11:00:00Z", "html_metadata_json": null, "is_ongoing": true, "user_permission": null, "created_at": "2023-11-28T16:30:32.190353Z", "edited_at": "2025-09-22T00:55:08.932036Z", "score_type": "peer_tournament", "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } ], "default_project": { "id": 2784, "type": "tournament", "name": "Chinese AI Chips", "slug": "chinese-ai-chips", "header_image": "https://metaculus-web-media.s3.amazonaws.com/cac-new_1.webp", "prize_pool": null, "start_date": "2023-12-06T12:00:33Z", "close_date": "2033-01-02T23:59:33Z", "forecasting_end_date": "2033-01-01T11:00:00Z", "html_metadata_json": null, "is_ongoing": true, "user_permission": null, "created_at": "2023-11-28T16:30:32.190353Z", "edited_at": "2025-09-22T00:55:08.932036Z", "score_type": "peer_tournament", "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" }, "category": [ { "id": 3701, "name": "Technology", "slug": "technology", "emoji": "⚙️", "description": "Technology", "type": "category" }, { "id": 3692, "name": "Computing and Math", "slug": "computing-and-math", "emoji": "💻", "description": "Computing and Math", "type": "category" } ], "question_series": [ { "id": 2343, "type": "question_series", "name": "AI in China", "slug": "ai-in-china", "header_image": "https://metaculus-web-media.s3.amazonaws.com/ai-china.png", "prize_pool": null, "start_date": null, "close_date": null, "forecasting_end_date": null, "html_metadata_json": null, "is_ongoing": true, "user_permission": "forecaster", "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2025-09-22T00:55:09.020170Z", "score_type": "relative_legacy_tournament", "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } ] }, "question": { "id": 20135, "title": "Will a Chinese firm market an EUV photolithography machine before 2033?", "created_at": "2023-11-24T10:38:48.175836Z", "open_time": "2023-12-06T11:00:00Z", "cp_reveal_time": "2023-12-20T11:00:00Z", "spot_scoring_time": "2023-12-20T11:00:00Z", "scheduled_resolve_time": "2033-01-01T11:00:00Z", "actual_resolve_time": null, "resolution_set_time": null, "scheduled_close_time": "2033-01-01T11:00:00Z", "actual_close_time": "2033-01-01T11:00:00Z", "type": "binary", "options": null, "group_variable": "", "status": "open", "possibilities": { "type": "binary" }, "resolution": null, "include_bots_in_aggregates": false, "question_weight": 1.0, "default_score_type": "peer", "default_aggregation_method": "recency_weighted", "label": "", "unit": "", "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "scaling": { "range_min": null, "range_max": null, "nominal_min": null, "nominal_max": null, "zero_point": null, "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "continuous_range": null }, "group_rank": null, "description": "In October 2022, the US instituted [new, wide-ranging export controls](https://www.csis.org/analysis/choking-chinas-access-future-ai) aimed at controlling a set of “chokepoint” technologies in the global semiconductor supply chain. China is one of the leading countries in AI, but AI advancements have historically run on compute, and China’s access to cost-competitive machine learning (ML) compute — the Chinese AI industry currently relies in large part on AI chips designed and fabricated abroad, and in particular Nvidia GPUs — is now in question.\n\nThe [US export controls](https://www.ecfr.gov/current/title-15/subtitle-B/chapter-VII/subchapter-C) prevent the export of high-performance AI chips to China, as well as the export of advanced tooling and materials used to make such chips. Still, one way for China to gain better access to cost-competitive ML compute is to indigenously produce high-quality AI chips. In 2015, China set out an ambitious plan aiming to bring the proportion of chips bought by Chinese firms that is domestically produced from 15% to 70% by 2025; as of 2021 [that number had gone up to 24%](https://asia.nikkei.com/Politics/International-relations/US-China-tensions/U.S.-tech-curbs-threaten-China-s-quest-for-chip-independence). With the October restrictions — and with US allies like Taiwan, Japan, South Korea, and the Netherlands enacting their own controls — Chinese firms now face substantial challenges in their attempts to indigenize semiconductor production.\n\nSome of the most important inputs into the chip-making process is semiconductor manufacturing equipment (SME) – highly complex and specialized tools essential to the chip-making process. (Other important inputs include specialized materials, software, and more generally capital, talent, and know-how.) Many of these tools — including most tools necessary for producing the most advanced chips — are prohibited from being exported to China.\n\nIn particular, the most important type of SME is photolithography machines, which are some of the most complex machinery ever made by humans. These machines expose a photomask (which holds the pattern that needs to be etched onto the silicon wafer) to light; the patterned light then reaches a photoresist material on the wafer. The exposed parts of the photoresist dissolve or harden, and the remaining parts can be removed to yield a mask for the next deposition or etching steps that form the chip’s circuits.\n\nThe most advanced type of photolithography machine used today is extreme ultraviolet (EUV) photolithography. As [Miller](https://www.christophermiller.net/semiconductors-1) writes, “EUV machines have multiple components that, on their own, constitute epically complex engineering challenges. Replicating just the laser in an EUV system requires perfectly identifying and assembling 457,329 parts. A single defect could cause debilitating delays or reliability problems.” The photolithography industry is dominated by the Dutch company ASML, which currently has a monopoly on EUV machines.\n\nChina’s major photolithography maker is Shanghai Micro Electronics Equipment (SMEE), founded in the early 2000s. SMEE [markets](http://www.smee.com.cn/eis.pub?service=homepageService&method=indexinfo&onclicknodeno=1_4_4_1) krypton fluoride (KrF) and argon fluoride (ArF) dry steppers for mature process nodes, but these either aren’t ready for mass production and/or are far behind (>15 years) Dutch and Japanese KrF and ArF machines in terms of resolution, throughput, and quality. SMEE does not market any more advanced argon fluoride immersion (ArFi) machines, let alone any EUV machines.\n\nFor more information, I have written a primer on [indigenously made Chinese AI chips](https://docs.google.com/document/d/1r-aLT67d2OQnuhZJZihkr0KxxB-PsDMKMut2flaYQ2c/edit?usp=sharing), intended to provide a useful introduction and reference for anyone interested in forecasting or learning more about China’s ability to indigenously produce AI chips. It explains some key concepts, provides an overview of the chip-making process, provides an overview of relevant export controls, and describes key inputs, context, and organizations relevant to understanding Chinese chip-making progress.", "resolution_criteria": "The question will resolve as **Yes** if a Chinese firm markets an extreme ultraviolet (EUV) photolithography machine on its website or credible sources report that a Chinese firm is selling EUV photolithography machines before January 1st, 2033. It will resolve as **No** otherwise.", "fine_print": "A “Chinese firm” is a firm headquartered in China. “China” includes mainland China, Hong Kong, and Macao, but not Taiwan. If the PRC annexes Taiwan, then annexed-Taiwan, including the chips being produced there, counts as being part of China.\n\nIt’s enough that the firm markets such a machine on either the English-language or Chinese-language version of its website. The product can be marketed either on a “product” page, or through a news release, or in any other form on its website, so long as it's accessible on the website.\n\nThe website or reporting from credible sources must mention both (a) that the product is an extreme ultraviolet (EUV) machine, and/or that it uses a 13.5 nm light, and (b) that it’s a lithography machine used for wafers or to fabricate semiconductors (or chips).\n\nIf the firm’s website lists an EUV machine prior to January 1st, 2033, but it’s then removed, the question still resolves as **Yes** (so long as Metaculus can confirm it).", "post_id": 20135, "aggregations": { "recency_weighted": { "history": [ { "start_time": 1758121278.242342, "end_time": 1759164197.409849, "forecaster_count": 36, "interval_lower_bounds": [ 0.6 ], "centers": [ 0.8 ], "interval_upper_bounds": [ 0.93 ] } ], "latest": { "start_time": 1758121278.242342, "end_time": 1759164197.409849, "forecaster_count": 36, "interval_lower_bounds": [ 0.6 ], "centers": [ 0.8 ], "interval_upper_bounds": [ 0.93 ], "forecast_values": [ 0.19999999999999996, 0.8 ], "means": [ 0.7063060238161953 ], "histogram": [ [ 0.0, 0.17249978086131545, 0.0, 0.0, 0.0, 0.09122295403992013, 0.0, 0.5928733813562592, 0.3325315572279212, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.3678794411714424, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.44110682688627867, 0.0, 0.0, 0.0, 0.0, 0.0, 0.014010486522538886, 0.0, 0.0, 0.5407300041287509, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.01831563888873418, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06833212923320346, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.6646021298955194, 0.0, 0.0, 0.0, 0.12593030301474023, 0.0, 0.0, 0.2999413040343729, 0.0, 1.5917533988848782, 0.0, 0.0, 0.13533528323661267, 0.041937583413620604, 0.7745776596494143, 0.0, 0.0, 0.0, 0.0, 1.0, 0.492383508225005, 0.0, 0.19376657685266277, 0.0, 0.1977322318500051, 0.0, 0.028709986438327752, 0.0, 2.3238610364852 ] ] }, "score_data": {}, "movement": null }, "unweighted": { "history": [], "latest": null, "score_data": {}, "movement": null }, "single_aggregation": { "history": [], "latest": null, "score_data": {}, "movement": null }, "metaculus_prediction": { "history": [ { "start_time": 1728289866.531852, "end_time": null, "forecaster_count": 29, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null } ], "latest": { "start_time": 1728289866.531852, "end_time": null, "forecaster_count": 29, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null, "forecast_values": [ 0.23901319618504013, 0.7609868038149599 ], "means": null, "histogram": null }, "score_data": {}, "movement": null } } }, "user_permission": "forecaster", "vote": { "score": 11, "user_vote": null }, "forecasts_count": 95, "key_factors": [], "is_current_content_translated": false, "description": "In October 2022, the US instituted [new, wide-ranging export controls](https://www.csis.org/analysis/choking-chinas-access-future-ai) aimed at controlling a set of “chokepoint” technologies in the global semiconductor supply chain. China is one of the leading countries in AI, but AI advancements have historically run on compute, and China’s access to cost-competitive machine learning (ML) compute — the Chinese AI industry currently relies in large part on AI chips designed and fabricated abroad, and in particular Nvidia GPUs — is now in question.\n\nThe [US export controls](https://www.ecfr.gov/current/title-15/subtitle-B/chapter-VII/subchapter-C) prevent the export of high-performance AI chips to China, as well as the export of advanced tooling and materials used to make such chips. Still, one way for China to gain better access to cost-competitive ML compute is to indigenously produce high-quality AI chips. In 2015, China set out an ambitious plan aiming to bring the proportion of chips bought by Chinese firms that is domestically produced from 15% to 70% by 2025; as of 2021 [that number had gone up to 24%](https://asia.nikkei.com/Politics/International-relations/US-China-tensions/U.S.-tech-curbs-threaten-China-s-quest-for-chip-independence). With the October restrictions — and with US allies like Taiwan, Japan, South Korea, and the Netherlands enacting their own controls — Chinese firms now face substantial challenges in their attempts to indigenize semiconductor production.\n\nSome of the most important inputs into the chip-making process is semiconductor manufacturing equipment (SME) – highly complex and specialized tools essential to the chip-making process. (Other important inputs include specialized materials, software, and more generally capital, talent, and know-how.) Many of these tools — including most tools necessary for producing the most advanced chips — are prohibited from being exported to China.\n\nIn particular, the most important type of SME is photolithography machines, which are some of the most complex machinery ever made by humans. These machines expose a photomask (which holds the pattern that needs to be etched onto the silicon wafer) to light; the patterned light then reaches a photoresist material on the wafer. The exposed parts of the photoresist dissolve or harden, and the remaining parts can be removed to yield a mask for the next deposition or etching steps that form the chip’s circuits.\n\nThe most advanced type of photolithography machine used today is extreme ultraviolet (EUV) photolithography. As [Miller](https://www.christophermiller.net/semiconductors-1) writes, “EUV machines have multiple components that, on their own, constitute epically complex engineering challenges. Replicating just the laser in an EUV system requires perfectly identifying and assembling 457,329 parts. A single defect could cause debilitating delays or reliability problems.” The photolithography industry is dominated by the Dutch company ASML, which currently has a monopoly on EUV machines.\n\nChina’s major photolithography maker is Shanghai Micro Electronics Equipment (SMEE), founded in the early 2000s. SMEE [markets](http://www.smee.com.cn/eis.pub?service=homepageService&method=indexinfo&onclicknodeno=1_4_4_1) krypton fluoride (KrF) and argon fluoride (ArF) dry steppers for mature process nodes, but these either aren’t ready for mass production and/or are far behind (>15 years) Dutch and Japanese KrF and ArF machines in terms of resolution, throughput, and quality. SMEE does not market any more advanced argon fluoride immersion (ArFi) machines, let alone any EUV machines.\n\nFor more information, I have written a primer on [indigenously made Chinese AI chips](https://docs.google.com/document/d/1r-aLT67d2OQnuhZJZihkr0KxxB-PsDMKMut2flaYQ2c/edit?usp=sharing), intended to provide a useful introduction and reference for anyone interested in forecasting or learning more about China’s ability to indigenously produce AI chips. It explains some key concepts, provides an overview of the chip-making process, provides an overview of relevant export controls, and describes key inputs, context, and organizations relevant to understanding Chinese chip-making progress." }, { "id": 20134, "title": "Will a Chinese-made ArFi lithography machine be used in volume production anywhere before 2028?", "short_title": "Chinese ArFi machine used at scale by 2026?", "url_title": "Chinese ArFi machine used at scale by 2026?", "slug": "chinese-arfi-machine-used-at-scale-by-2026", "author_id": 111541, "author_username": "DerMannOhneEigenschaften", "coauthors": [], "created_at": "2023-11-24T10:35:43.883430Z", "published_at": "2023-12-06T11:00:00Z", "edited_at": "2025-09-10T10:17:24.709498Z", "curation_status": "approved", "curation_status_updated_at": "2023-12-06T11:00:00Z", "comment_count": 0, "status": "open", "resolved": false, "actual_close_time": null, "scheduled_close_time": "2028-01-01T11:00:00Z", "scheduled_resolve_time": "2028-02-01T11:00:00Z", "actual_resolve_time": null, "open_time": "2023-12-06T11:00:00Z", "nr_forecasters": 25, "html_metadata_json": null, "projects": { "site_main": [ { "id": 144, "type": "site_main", "name": "Metaculus Community", "slug": null, "header_image": null, "prize_pool": null, "start_date": null, "close_date": null, "forecasting_end_date": null, "html_metadata_json": null, "is_ongoing": null, "user_permission": "forecaster", "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2025-07-18T17:28:18.838588Z", "score_type": null, "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } ], "tournament": [ { "id": 2784, "type": "tournament", "name": "Chinese AI Chips", "slug": "chinese-ai-chips", "header_image": "https://metaculus-web-media.s3.amazonaws.com/cac-new_1.webp", "prize_pool": null, "start_date": "2023-12-06T12:00:33Z", "close_date": "2033-01-02T23:59:33Z", "forecasting_end_date": "2033-01-01T11:00:00Z", "html_metadata_json": null, "is_ongoing": true, "user_permission": null, "created_at": "2023-11-28T16:30:32.190353Z", "edited_at": "2025-09-22T00:55:08.932036Z", "score_type": "peer_tournament", "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } ], "default_project": { "id": 2784, "type": "tournament", "name": "Chinese AI Chips", "slug": "chinese-ai-chips", "header_image": "https://metaculus-web-media.s3.amazonaws.com/cac-new_1.webp", "prize_pool": null, "start_date": "2023-12-06T12:00:33Z", "close_date": "2033-01-02T23:59:33Z", "forecasting_end_date": "2033-01-01T11:00:00Z", "html_metadata_json": null, "is_ongoing": true, "user_permission": null, "created_at": "2023-11-28T16:30:32.190353Z", "edited_at": "2025-09-22T00:55:08.932036Z", "score_type": "peer_tournament", "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" }, "category": [ { "id": 3701, "name": "Technology", "slug": "technology", "emoji": "⚙️", "description": "Technology", "type": "category" }, { "id": 3692, "name": "Computing and Math", "slug": "computing-and-math", "emoji": "💻", "description": "Computing and Math", "type": "category" } ], "question_series": [ { "id": 2343, "type": "question_series", "name": "AI in China", "slug": "ai-in-china", "header_image": "https://metaculus-web-media.s3.amazonaws.com/ai-china.png", "prize_pool": null, "start_date": null, "close_date": null, "forecasting_end_date": null, "html_metadata_json": null, "is_ongoing": true, "user_permission": "forecaster", "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2025-09-22T00:55:09.020170Z", "score_type": "relative_legacy_tournament", "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } ] }, "question": { "id": 20134, "title": "Will a Chinese-made ArFi lithography machine be used in volume production anywhere before 2028?", "created_at": "2023-11-24T10:35:43.883430Z", "open_time": "2023-12-06T11:00:00Z", "cp_reveal_time": "2023-12-20T11:00:00Z", "spot_scoring_time": "2023-12-20T11:00:00Z", "scheduled_resolve_time": "2028-02-01T11:00:00Z", "actual_resolve_time": null, "resolution_set_time": null, "scheduled_close_time": "2028-01-01T11:00:00Z", "actual_close_time": "2028-01-01T11:00:00Z", "type": "binary", "options": null, "group_variable": "", "status": "open", "possibilities": { "type": "binary" }, "resolution": null, "include_bots_in_aggregates": false, "question_weight": 1.0, "default_score_type": "peer", "default_aggregation_method": "recency_weighted", "label": "", "unit": "", "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "scaling": { "range_min": null, "range_max": null, "nominal_min": null, "nominal_max": null, "zero_point": null, "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "continuous_range": null }, "group_rank": null, "description": "In October 2022, the US instituted [new, wide-ranging export controls](https://www.csis.org/analysis/choking-chinas-access-future-ai) aimed at controlling a set of “chokepoint” technologies in the global semiconductor supply chain. China is one of the leading countries in AI, but AI advancements have historically run on compute, and China’s access to cost-competitive machine learning (ML) compute — the Chinese AI industry currently relies in large part on AI chips designed and fabricated abroad, and in particular Nvidia GPUs — is now in question.\n\nThe [US export controls](https://www.ecfr.gov/current/title-15/subtitle-B/chapter-VII/subchapter-C) prevent the export of high-performance AI chips to China, as well as the export of advanced tooling and materials used to make such chips. Still, one way for China to gain better access to cost-competitive ML compute is to indigenously produce high-quality AI chips. In 2015, China set out an ambitious plan aiming to bring the proportion of chips bought by Chinese firms that is domestically produced from 15% to 70% by 2025; as of 2021 [that number had gone up to 24%](https://asia.nikkei.com/Politics/International-relations/US-China-tensions/U.S.-tech-curbs-threaten-China-s-quest-for-chip-independence). With the October restrictions — and with US allies like Taiwan, Japan, South Korea, and the Netherlands enacting their own controls — Chinese firms now face substantial challenges in their attempts to indigenize semiconductor production.\n\nSome of the most important inputs into the chip-making process is semiconductor manufacturing equipment (SME) – highly complex and specialized tools essential to the chip-making process. (Other important inputs include specialized materials, software, and more generally capital, talent, and know-how.) Many of these tools — including most tools necessary for producing the most advanced chips — are prohibited from being exported to China.\n\nIn particular, the most important type of SME is photolithography machines, which are some of the most complex machinery ever made by humans. These machines expose a photomask (which holds the pattern that needs to be etched onto the silicon wafer) to light; the patterned light then reaches a photoresist material on the wafer. The exposed parts of the photoresist dissolve or harden, and the remaining parts can be removed to yield a mask for the next deposition or etching steps that form the chip’s circuits. The photolithography industry is dominated by the Dutch company ASML.\n\nChina’s major [photolithography](https://en.wikipedia.org/wiki/Photolithography) maker is Shanghai Micro Electronics Equipment (SMEE), founded in the early 2000s. SMEE [markets](http://www.smee.com.cn/eis.pub?service=homepageService&method=indexinfo&onclicknodeno=1_4_4_1) krypton fluoride (KrF) and argon fluoride (ArF) dry steppers for mature process nodes, but these either aren’t ready for mass production and/or are far behind (>15 years) Dutch and Japanese KrF and ArF machines in terms of resolution, throughput, and quality. SMEE does not market any more advanced argon fluoride immersion (ArFi) machines, though it is [reportedly](https://www.tomshardware.com/news/chinese-lithography-firm-to-debut-28nm-capable-scanner-report) “on track to reveal its first scanner capable of producing chips on a 28nm process technology by the end of 2023”.\n\nFor more information, I have written a primer on [indigenously made Chinese AI chips](https://docs.google.com/document/d/1r-aLT67d2OQnuhZJZihkr0KxxB-PsDMKMut2flaYQ2c/edit?usp=sharing), intended to provide a useful introduction and reference for anyone interested in forecasting or learning more about China’s ability to indigenously produce AI chips. It explains some key concepts, provides an overview of the chip-making process, provides an overview of relevant export controls, and describes key inputs, context, and organizations relevant to understanding Chinese chip-making progress.", "resolution_criteria": "The question will resolve as **Yes** if an argon fluoride immersion (ArFi) photolithography machine made by a Chinese firm is used in volume production anywhere before January 1st, 2028. The question will resolve as **No** otherwise.\n\nThe question resolves as **Yes** if and only if credible sources report that a Chinese-made ArFi machine has been used in volume production, and/or if a semiconductor fabrication company itself claims it produces 50,000 ore more wafers per month using a Chinese-made ArFi machine. The reports must mention that the fab has achieved “(high) volume production” or “mass production” using the machine.", "fine_print": "A “Chinese firm” is a firm headquartered in China. “China” includes mainland China, Hong Kong, and Macao, but not Taiwan. If the PRC annexes Taiwan, then annexed-Taiwan, including the chips being produced there, counts as being part of China.\n\nThe reports have to be published prior to 2028, or concern production that happened prior to 2028 in reporting available before February 1, 2028.", "post_id": 20134, "aggregations": { "recency_weighted": { "history": [ { "start_time": 1758141710.388138, "end_time": 1759181586.545299, "forecaster_count": 14, "interval_lower_bounds": [ 0.72 ], "centers": [ 0.8 ], "interval_upper_bounds": [ 0.87 ] } ], "latest": { "start_time": 1758141710.388138, "end_time": 1759181586.545299, "forecaster_count": 14, "interval_lower_bounds": [ 0.72 ], "centers": [ 0.8 ], "interval_upper_bounds": [ 0.87 ], "forecast_values": [ 0.19999999999999996, 0.8 ], "means": [ 0.8083159857982065 ], "histogram": [ [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0644634174904052, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.09754477024200368, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.7313528357211682, 0.0, 0.13404139897044384, 0.0, 1.0, 0.0, 0.0, 0.27467474026865935, 0.0, 0.0, 0.0, 0.0, 1.1588593721215967, 0.560245764355437, 0.0, 0.0, 0.0, 0.0, 0.0, 0.8727500189273638, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 1.1300722830890126 ] ] }, "score_data": {}, "movement": null }, "unweighted": { "history": [], "latest": null, "score_data": {}, "movement": null }, "single_aggregation": { "history": [], "latest": null, "score_data": {}, "movement": null }, "metaculus_prediction": { "history": [ { "start_time": 1728287976.204529, "end_time": null, "forecaster_count": 17, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null } ], "latest": { "start_time": 1728287976.204529, "end_time": null, "forecaster_count": 17, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null, "forecast_values": [ 0.47525889022039014, 0.5247411097796099 ], "means": null, "histogram": null }, "score_data": {}, "movement": null } } }, "user_permission": "forecaster", "vote": { "score": 7, "user_vote": null }, "forecasts_count": 82, "key_factors": [], "is_current_content_translated": false, "description": "In October 2022, the US instituted [new, wide-ranging export controls](https://www.csis.org/analysis/choking-chinas-access-future-ai) aimed at controlling a set of “chokepoint” technologies in the global semiconductor supply chain. China is one of the leading countries in AI, but AI advancements have historically run on compute, and China’s access to cost-competitive machine learning (ML) compute — the Chinese AI industry currently relies in large part on AI chips designed and fabricated abroad, and in particular Nvidia GPUs — is now in question.\n\nThe [US export controls](https://www.ecfr.gov/current/title-15/subtitle-B/chapter-VII/subchapter-C) prevent the export of high-performance AI chips to China, as well as the export of advanced tooling and materials used to make such chips. Still, one way for China to gain better access to cost-competitive ML compute is to indigenously produce high-quality AI chips. In 2015, China set out an ambitious plan aiming to bring the proportion of chips bought by Chinese firms that is domestically produced from 15% to 70% by 2025; as of 2021 [that number had gone up to 24%](https://asia.nikkei.com/Politics/International-relations/US-China-tensions/U.S.-tech-curbs-threaten-China-s-quest-for-chip-independence). With the October restrictions — and with US allies like Taiwan, Japan, South Korea, and the Netherlands enacting their own controls — Chinese firms now face substantial challenges in their attempts to indigenize semiconductor production.\n\nSome of the most important inputs into the chip-making process is semiconductor manufacturing equipment (SME) – highly complex and specialized tools essential to the chip-making process. (Other important inputs include specialized materials, software, and more generally capital, talent, and know-how.) Many of these tools — including most tools necessary for producing the most advanced chips — are prohibited from being exported to China.\n\nIn particular, the most important type of SME is photolithography machines, which are some of the most complex machinery ever made by humans. These machines expose a photomask (which holds the pattern that needs to be etched onto the silicon wafer) to light; the patterned light then reaches a photoresist material on the wafer. The exposed parts of the photoresist dissolve or harden, and the remaining parts can be removed to yield a mask for the next deposition or etching steps that form the chip’s circuits. The photolithography industry is dominated by the Dutch company ASML.\n\nChina’s major [photolithography](https://en.wikipedia.org/wiki/Photolithography) maker is Shanghai Micro Electronics Equipment (SMEE), founded in the early 2000s. SMEE [markets](http://www.smee.com.cn/eis.pub?service=homepageService&method=indexinfo&onclicknodeno=1_4_4_1) krypton fluoride (KrF) and argon fluoride (ArF) dry steppers for mature process nodes, but these either aren’t ready for mass production and/or are far behind (>15 years) Dutch and Japanese KrF and ArF machines in terms of resolution, throughput, and quality. SMEE does not market any more advanced argon fluoride immersion (ArFi) machines, though it is [reportedly](https://www.tomshardware.com/news/chinese-lithography-firm-to-debut-28nm-capable-scanner-report) “on track to reveal its first scanner capable of producing chips on a 28nm process technology by the end of 2023”.\n\nFor more information, I have written a primer on [indigenously made Chinese AI chips](https://docs.google.com/document/d/1r-aLT67d2OQnuhZJZihkr0KxxB-PsDMKMut2flaYQ2c/edit?usp=sharing), intended to provide a useful introduction and reference for anyone interested in forecasting or learning more about China’s ability to indigenously produce AI chips. It explains some key concepts, provides an overview of the chip-making process, provides an overview of relevant export controls, and describes key inputs, context, and organizations relevant to understanding Chinese chip-making progress." }, { "id": 20133, "title": "Will a Chinese firm market an ArFi photolithography machine before 2025?", "short_title": "Chinese ArFi machine marketed by 2025?", "url_title": "Chinese ArFi machine marketed by 2025?", "slug": "chinese-arfi-machine-marketed-by-2025", "author_id": 111541, "author_username": "DerMannOhneEigenschaften", "coauthors": [], "created_at": "2023-11-24T10:29:13.814960Z", "published_at": "2023-12-06T11:00:00Z", "edited_at": "2025-09-05T17:28:50.326369Z", "curation_status": "approved", "curation_status_updated_at": "2023-12-06T11:00:00Z", "comment_count": 9, "status": "resolved", "resolved": true, "actual_close_time": "2025-01-01T11:00:00Z", "scheduled_close_time": "2025-01-01T11:00:00Z", "scheduled_resolve_time": "2025-01-01T11:00:00Z", "actual_resolve_time": "2025-01-10T00:44:00Z", "open_time": "2023-12-06T11:00:00Z", "nr_forecasters": 33, "html_metadata_json": null, "projects": { "leaderboard_tag": [ { "id": 32590, "name": "2021-2025 Leaderboard", "slug": "2021_2025_leaderboard", "type": "leaderboard_tag" } ], "site_main": [ { "id": 144, "type": "site_main", "name": "Metaculus Community", "slug": null, "header_image": null, "prize_pool": null, "start_date": null, "close_date": null, "forecasting_end_date": null, "html_metadata_json": null, "is_ongoing": null, "user_permission": "forecaster", "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2025-07-18T17:28:18.838588Z", "score_type": null, "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } ], "tournament": [ { "id": 2784, "type": "tournament", "name": "Chinese AI Chips", "slug": "chinese-ai-chips", "header_image": "https://metaculus-web-media.s3.amazonaws.com/cac-new_1.webp", "prize_pool": null, "start_date": "2023-12-06T12:00:33Z", "close_date": "2033-01-02T23:59:33Z", "forecasting_end_date": "2033-01-01T11:00:00Z", "html_metadata_json": null, "is_ongoing": true, "user_permission": null, "created_at": "2023-11-28T16:30:32.190353Z", "edited_at": "2025-09-22T00:55:08.932036Z", "score_type": "peer_tournament", "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } ], "default_project": { "id": 2784, "type": "tournament", "name": "Chinese AI Chips", "slug": "chinese-ai-chips", "header_image": "https://metaculus-web-media.s3.amazonaws.com/cac-new_1.webp", "prize_pool": null, "start_date": "2023-12-06T12:00:33Z", "close_date": "2033-01-02T23:59:33Z", "forecasting_end_date": "2033-01-01T11:00:00Z", "html_metadata_json": null, "is_ongoing": true, "user_permission": null, "created_at": "2023-11-28T16:30:32.190353Z", "edited_at": "2025-09-22T00:55:08.932036Z", "score_type": "peer_tournament", "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" }, "category": [ { "id": 3701, "name": "Technology", "slug": "technology", "emoji": "⚙️", "description": "Technology", "type": "category" }, { "id": 3692, "name": "Computing and Math", "slug": "computing-and-math", "emoji": "💻", "description": "Computing and Math", "type": "category" } ], "question_series": [ { "id": 2343, "type": "question_series", "name": "AI in China", "slug": "ai-in-china", "header_image": "https://metaculus-web-media.s3.amazonaws.com/ai-china.png", "prize_pool": null, "start_date": null, "close_date": null, "forecasting_end_date": null, "html_metadata_json": null, "is_ongoing": true, "user_permission": "forecaster", "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2025-09-22T00:55:09.020170Z", "score_type": "relative_legacy_tournament", "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } ] }, "question": { "id": 20133, "title": "Will a Chinese firm market an ArFi photolithography machine before 2025?", "created_at": "2023-11-24T10:29:13.814960Z", "open_time": "2023-12-06T11:00:00Z", "cp_reveal_time": "2023-12-20T11:00:00Z", "spot_scoring_time": "2023-12-20T11:00:00Z", "scheduled_resolve_time": "2025-01-01T11:00:00Z", "actual_resolve_time": "2025-01-10T00:44:00Z", "resolution_set_time": "2025-01-10T00:51:13.022598Z", "scheduled_close_time": "2025-01-01T11:00:00Z", "actual_close_time": "2025-01-01T11:00:00Z", "type": "binary", "options": null, "group_variable": "", "status": "resolved", "possibilities": { "type": "binary" }, "resolution": "no", "include_bots_in_aggregates": false, "question_weight": 1.0, "default_score_type": "peer", "default_aggregation_method": "recency_weighted", "label": "", "unit": "", "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "scaling": { "range_min": null, "range_max": null, "nominal_min": null, "nominal_max": null, "zero_point": null, "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "continuous_range": null }, "group_rank": null, "description": "In October 2022, the US instituted [new, wide-ranging export controls](https://www.csis.org/analysis/choking-chinas-access-future-ai) aimed at controlling a set of “chokepoint” technologies in the global semiconductor supply chain. China is one of the leading countries in AI, but AI advancements have historically run on compute, and China’s access to cost-competitive machine learning (ML) compute — the Chinese AI industry currently relies in large part on AI chips designed and fabricated abroad, and in particular Nvidia GPUs — is now in question.\n\nThe [US export controls](https://www.ecfr.gov/current/title-15/subtitle-B/chapter-VII/subchapter-C) prevent the export of high-performance AI chips to China, as well as the export of advanced tooling and materials used to make such chips. Still, one way for China to gain better access to cost-competitive ML compute is to indigenously produce high-quality AI chips. In 2015, China set out an ambitious plan aiming to bring the proportion of chips bought by Chinese firms that is domestically produced from 15% to 70% by 2025; as of 2021 [that number had gone up to 24%](https://asia.nikkei.com/Politics/International-relations/US-China-tensions/U.S.-tech-curbs-threaten-China-s-quest-for-chip-independence). With the October restrictions — and with US allies like Taiwan, Japan, South Korea, and the Netherlands enacting their own controls — Chinese firms now face substantial challenges in their attempts to indigenize semiconductor production.\n\nSome of the most important inputs into the chip-making process is semiconductor manufacturing equipment (SME) – highly complex and specialized tools essential to the chip-making process. (Other important inputs include specialized materials, software, and more generally capital, talent, and know-how.) Many of these tools — including most tools necessary for producing the most advanced chips — are prohibited from being exported to China.\n\nIn particular, the most important type of SME is photolithography machines, which are some of the most complex machinery ever made by humans. These machines expose a photomask (which holds the pattern that needs to be etched onto the silicon wafer) to light; the patterned light then reaches a photoresist material on the wafer. The exposed parts of the photoresist dissolve or harden, and the remaining parts can be removed to yield a mask for the next deposition or etching steps that form the chip’s circuits. The photolithography industry is dominated by the Dutch company ASML.\n\nChina’s major [photolithography](https://en.wikipedia.org/wiki/Photolithography) maker is Shanghai Micro Electronics Equipment (SMEE), founded in the early 2000s. SMEE [markets](http://www.smee.com.cn/eis.pub?service=homepageService&method=indexinfo&onclicknodeno=1_4_4_1) krypton fluoride (KrF) and argon fluoride (ArF) dry steppers for mature process nodes, but these either aren’t ready for mass production and/or are far behind (>15 years) Dutch and Japanese KrF and ArF machines in terms of resolution, throughput, and quality. SMEE does not market any more advanced argon fluoride immersion (ArFi) machines, though it is [reportedly](https://www.tomshardware.com/news/chinese-lithography-firm-to-debut-28nm-capable-scanner-report) “on track to reveal its first scanner capable of producing chips on a 28nm process technology by the end of 2023”.\n\nFor more information, I have written a primer on [indigenously made Chinese AI chips](https://docs.google.com/document/d/1r-aLT67d2OQnuhZJZihkr0KxxB-PsDMKMut2flaYQ2c/edit?usp=sharing), intended to provide a useful introduction and reference for anyone interested in forecasting or learning more about China’s ability to indigenously produce AI chips. It explains some key concepts, provides an overview of the chip-making process, provides an overview of relevant export controls, and describes key inputs, context, and organizations relevant to understanding Chinese chip-making progress.", "resolution_criteria": "The question will resolve as **Yes** if a Chinese firm markets an argon fluoride immersion (ArFi) photolithography machine on its website or credible sources report that a Chinese firm is selling ArFi photolithography machines before January 1st, 2025. The question will resolve as **No** otherwise.", "fine_print": "A “Chinese firm” is a firm headquartered in China. “China” includes mainland China, Hong Kong, and Macao (but not Taiwan). If the PRC annexes Taiwan, then annexed-Taiwan, including the chips being produced there, counts as being part of China.\n\nIt’s enough that the firm markets such a machine on either the English-language or Chinese-language version of its website. The product can be marketed either on a “product” page, or through a news release, or in any other form on its website, so long as it's accessible on the website.\n\nThe website or reporting must mention both (a) that the product is an argon fluoride (ArF or ArFi) machine, and/or that it uses a 193 nm light, and (b) that it’s a lithography machine used for wafers or to fabricate semiconductors (or chips).\n\nIf the firm’s website lists an ArFi machine prior to January 1st, 2025, but it’s then removed, the question still resolves as **Yes** (so long as Metaculus can confirm it).", "post_id": 20133, "aggregations": { "recency_weighted": { "history": [ { "start_time": 1735726608.279532, "end_time": null, "forecaster_count": 31, "interval_lower_bounds": [ 0.001 ], "centers": [ 0.01 ], "interval_upper_bounds": [ 0.07 ] } ], "latest": { "start_time": 1735726608.279532, "end_time": null, "forecaster_count": 31, "interval_lower_bounds": [ 0.001 ], "centers": [ 0.01 ], "interval_upper_bounds": [ 0.07 ], "forecast_values": [ 0.99, 0.01 ], "means": [ 0.08697377799920539 ], "histogram": [ [ 4.115335542517827, 2.2665452090358746, 0.0, 0.5123307444416587, 0.2358265416482139, 0.03573243764150962, 0.0, 0.10527918291777463, 0.0, 0.0, 0.46211900267680595, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0902215738421042, 0.14359465189107992, 0.0, 0.010381142261586365, 0.0, 0.0, 0.0, 0.29853580031441634, 0.07670684254184146, 0.0646131558445074, 0.0, 0.1405470336200478, 0.0, 0.26577008775178024, 0.0, 0.0, 0.33432944568041745, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20851081619995687, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.04392740981872405, 0.0, 0.0, 0.0, 0.0, 0.05382522520245611, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1836394386094319, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 ] ] }, "score_data": { "baseline_score": 55.60052495556688, "peer_score": 31.010478680832012, "coverage": 0.9990152833777399, "relative_legacy_score": 0.0, "weighted_coverage": 0.9990152833777399, "spot_peer_score": 72.75329954261082, "spot_baseline_score": 58.496250072115615, "baseline_archived_score": 55.60052495556688, "peer_archived_score": 31.010478680832012, "relative_legacy_archived_score": 0.0, "spot_peer_archived_score": 72.75329954261082, "spot_baseline_archived_score": 58.496250072115615 }, "movement": null }, "unweighted": { "history": [], "latest": null, "score_data": {}, "movement": null }, "single_aggregation": { "history": [], "latest": null, "score_data": {}, "movement": null }, "metaculus_prediction": { "history": [ { "start_time": 1728288006.222721, "end_time": null, "forecaster_count": 28, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null } ], "latest": { "start_time": 1728288006.222721, "end_time": null, "forecaster_count": 28, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null, "forecast_values": [ 0.9588489610548505, 0.04115103894514952 ], "means": null, "histogram": null }, "score_data": {}, "movement": null } } }, "user_permission": "forecaster", "vote": { "score": 9, "user_vote": null }, "forecasts_count": 112, "key_factors": [], "is_current_content_translated": false, "description": "In October 2022, the US instituted [new, wide-ranging export controls](https://www.csis.org/analysis/choking-chinas-access-future-ai) aimed at controlling a set of “chokepoint” technologies in the global semiconductor supply chain. China is one of the leading countries in AI, but AI advancements have historically run on compute, and China’s access to cost-competitive machine learning (ML) compute — the Chinese AI industry currently relies in large part on AI chips designed and fabricated abroad, and in particular Nvidia GPUs — is now in question.\n\nThe [US export controls](https://www.ecfr.gov/current/title-15/subtitle-B/chapter-VII/subchapter-C) prevent the export of high-performance AI chips to China, as well as the export of advanced tooling and materials used to make such chips. Still, one way for China to gain better access to cost-competitive ML compute is to indigenously produce high-quality AI chips. In 2015, China set out an ambitious plan aiming to bring the proportion of chips bought by Chinese firms that is domestically produced from 15% to 70% by 2025; as of 2021 [that number had gone up to 24%](https://asia.nikkei.com/Politics/International-relations/US-China-tensions/U.S.-tech-curbs-threaten-China-s-quest-for-chip-independence). With the October restrictions — and with US allies like Taiwan, Japan, South Korea, and the Netherlands enacting their own controls — Chinese firms now face substantial challenges in their attempts to indigenize semiconductor production.\n\nSome of the most important inputs into the chip-making process is semiconductor manufacturing equipment (SME) – highly complex and specialized tools essential to the chip-making process. (Other important inputs include specialized materials, software, and more generally capital, talent, and know-how.) Many of these tools — including most tools necessary for producing the most advanced chips — are prohibited from being exported to China.\n\nIn particular, the most important type of SME is photolithography machines, which are some of the most complex machinery ever made by humans. These machines expose a photomask (which holds the pattern that needs to be etched onto the silicon wafer) to light; the patterned light then reaches a photoresist material on the wafer. The exposed parts of the photoresist dissolve or harden, and the remaining parts can be removed to yield a mask for the next deposition or etching steps that form the chip’s circuits. The photolithography industry is dominated by the Dutch company ASML.\n\nChina’s major [photolithography](https://en.wikipedia.org/wiki/Photolithography) maker is Shanghai Micro Electronics Equipment (SMEE), founded in the early 2000s. SMEE [markets](http://www.smee.com.cn/eis.pub?service=homepageService&method=indexinfo&onclicknodeno=1_4_4_1) krypton fluoride (KrF) and argon fluoride (ArF) dry steppers for mature process nodes, but these either aren’t ready for mass production and/or are far behind (>15 years) Dutch and Japanese KrF and ArF machines in terms of resolution, throughput, and quality. SMEE does not market any more advanced argon fluoride immersion (ArFi) machines, though it is [reportedly](https://www.tomshardware.com/news/chinese-lithography-firm-to-debut-28nm-capable-scanner-report) “on track to reveal its first scanner capable of producing chips on a 28nm process technology by the end of 2023”.\n\nFor more information, I have written a primer on [indigenously made Chinese AI chips](https://docs.google.com/document/d/1r-aLT67d2OQnuhZJZihkr0KxxB-PsDMKMut2flaYQ2c/edit?usp=sharing), intended to provide a useful introduction and reference for anyone interested in forecasting or learning more about China’s ability to indigenously produce AI chips. It explains some key concepts, provides an overview of the chip-making process, provides an overview of relevant export controls, and describes key inputs, context, and organizations relevant to understanding Chinese chip-making progress." }, { "id": 20127, "title": "Will any country score in the top tier overall in the next Global Health Security Index?", "short_title": "Top Tier Overall Score in next GHS Index?", "url_title": "Top Tier Overall Score in next GHS Index?", "slug": "top-tier-overall-score-in-next-ghs-index", "author_id": 119839, "author_username": "DaAdCh", "coauthors": [], "created_at": "2023-11-23T21:13:29.634144Z", "published_at": "2023-12-30T20:30:00Z", "edited_at": "2025-09-05T17:28:56.066204Z", "curation_status": "approved", "curation_status_updated_at": "2023-12-30T20:30:00Z", "comment_count": 5, "status": "closed", "resolved": false, "actual_close_time": "2024-12-31T12:00:00Z", "scheduled_close_time": "2024-12-31T12:00:00Z", "scheduled_resolve_time": "2025-12-31T12:00:00Z", "actual_resolve_time": null, "open_time": "2023-12-30T20:30:00Z", "nr_forecasters": 14, "html_metadata_json": null, "projects": { "leaderboard_tag": [ { "id": 32590, "name": "2021-2025 Leaderboard", "slug": "2021_2025_leaderboard", "type": "leaderboard_tag" } ], "topic": [ { "id": 15865, "name": "Health & Pandemics", "slug": "biosecurity", "emoji": "🧬", "type": "topic" } ], "site_main": [ { "id": 144, "type": "site_main", "name": "Metaculus Community", "slug": null, "header_image": null, "prize_pool": null, "start_date": null, "close_date": null, "forecasting_end_date": null, "html_metadata_json": null, "is_ongoing": null, "user_permission": "forecaster", "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2025-07-18T17:28:18.838588Z", "score_type": null, "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } ], "default_project": { "id": 144, "type": "site_main", "name": "Metaculus Community", "slug": null, "header_image": null, "prize_pool": null, "start_date": null, "close_date": null, "forecasting_end_date": null, "html_metadata_json": null, "is_ongoing": null, "user_permission": "forecaster", "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2025-07-18T17:28:18.838588Z", "score_type": null, "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" }, "category": [ { "id": 3698, "name": "Economy & Business", "slug": "economy-business", "emoji": "💼", "description": "Economy & Business", "type": "category" }, { "id": 3691, "name": "Health & Pandemics", "slug": "health-pandemics", "emoji": "🦠", "description": "Health & Pandemics", "type": "category" } ] }, "question": { "id": 20127, "title": "Will any country score in the top tier overall in the next Global Health Security Index?", "created_at": "2023-11-23T21:13:29.634144Z", "open_time": "2023-12-30T20:30:00Z", "cp_reveal_time": "2024-01-02T20:30:00Z", "spot_scoring_time": "2024-01-02T20:30:00Z", "scheduled_resolve_time": "2025-12-31T12:00:00Z", "actual_resolve_time": null, "resolution_set_time": null, "scheduled_close_time": "2024-12-31T12:00:00Z", "actual_close_time": "2024-12-31T12:00:00Z", "type": "binary", "options": [], "group_variable": "", "status": "closed", "possibilities": { "type": "binary" }, "resolution": null, "include_bots_in_aggregates": false, "question_weight": 1.0, "default_score_type": "peer", "default_aggregation_method": "recency_weighted", "label": "", "unit": "", "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "scaling": { "range_min": null, "range_max": null, "nominal_min": null, "nominal_max": null, "zero_point": null, "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "continuous_range": null }, "group_rank": null, "description": "The [Global Health Security (GHS) Index](https://www.ghsindex.org/) is an assessment and benchmarking of health security and related capabilities across 195 countries. The GHS Index was first launched in October 2019.\n\nIn [the latest report](https://www.ghsindex.org/wp-content/uploads/2021/12/2021_GHSindexFullReport_Final.pdf), published in December 2021, countries were assessed across 6 categories, 37 indicators, and 171 questions using publicly available information. The GHS Index benchmarks health security in the context of other factors critical to fighting outbreaks, such as political and security risks, the broader strength of the health system, and country adherence to global norms. No country scored in the top tier overall; the United States of America was top with a score of 75.9 (out of 100), 4.2 below the threshold for top tier.\n\nThe GHS Index is designed to inform leaders of the foundational elements that are necessary to prepare their countries for future outbreaks and where they should prioritise planning and invest durable funding. By assessing these capacities every 2-3 years, the GHS Index stimulates political will and action to prioritise addressing these gaps. The next GHS Index report is due in 2024 but could be later. Arguably, it will be the first report since the Covid-19 epidemic has reduced in intensity across the globe.", "resolution_criteria": "This question will resolve Yes if at least one country achieves an overall score in the top tier category in the next [Global Health Security (GHS) Index](https://www.ghsindex.org/) report due in 2024 (or later). The top tier category, as described in the 2021 GHS Index report, is an overall score of 80.1 or more (out of 100). If the next report is not published before January 1, 2026 or the scoring system changes, this question will be annulled.", "fine_print": "", "post_id": 20127, "aggregations": { "recency_weighted": { "history": [ { "start_time": 1735403814.247364, "end_time": null, "forecaster_count": 13, "interval_lower_bounds": [ 0.19 ], "centers": [ 0.2 ], "interval_upper_bounds": [ 0.2 ] } ], "latest": { "start_time": 1735403814.247364, "end_time": null, "forecaster_count": 13, "interval_lower_bounds": [ 0.19 ], "centers": [ 0.2 ], "interval_upper_bounds": [ 0.2 ], "forecast_values": [ 0.8, 0.2 ], "means": [ 0.24533765207684616 ], "histogram": [ [ 0.0, 0.0, 0.0, 0.0, 0.25423829106281975, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.6419315407681068, 0.0, 0.0, 0.3147232704804096, 0.0, 1.0, 2.660640686416884, 0.07386240743899689, 0.0, 0.0, 0.0, 0.1535851000440937, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.11176713620916234, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.5457734721688206, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 ] ] }, "score_data": {}, "movement": null }, "unweighted": { "history": [], "latest": null, "score_data": {}, "movement": null }, "single_aggregation": { "history": [], "latest": null, "score_data": {}, "movement": null }, "metaculus_prediction": { "history": [ { "start_time": 1728287055.672188, "end_time": null, "forecaster_count": 9, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null } ], "latest": { "start_time": 1728287055.672188, "end_time": null, "forecaster_count": 9, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null, "forecast_values": [ 0.9341893515167263, 0.06581064848327366 ], "means": null, "histogram": null }, "score_data": {}, "movement": null } } }, "user_permission": "forecaster", "vote": { "score": 3, "user_vote": null }, "forecasts_count": 35, "key_factors": [], "is_current_content_translated": false, "description": "The [Global Health Security (GHS) Index](https://www.ghsindex.org/) is an assessment and benchmarking of health security and related capabilities across 195 countries. The GHS Index was first launched in October 2019.\n\nIn [the latest report](https://www.ghsindex.org/wp-content/uploads/2021/12/2021_GHSindexFullReport_Final.pdf), published in December 2021, countries were assessed across 6 categories, 37 indicators, and 171 questions using publicly available information. The GHS Index benchmarks health security in the context of other factors critical to fighting outbreaks, such as political and security risks, the broader strength of the health system, and country adherence to global norms. No country scored in the top tier overall; the United States of America was top with a score of 75.9 (out of 100), 4.2 below the threshold for top tier.\n\nThe GHS Index is designed to inform leaders of the foundational elements that are necessary to prepare their countries for future outbreaks and where they should prioritise planning and invest durable funding. By assessing these capacities every 2-3 years, the GHS Index stimulates political will and action to prioritise addressing these gaps. The next GHS Index report is due in 2024 but could be later. Arguably, it will be the first report since the Covid-19 epidemic has reduced in intensity across the globe." }, { "id": 20118, "title": "Will 20 million people follow a new, AI-created religion by the end of 2035?", "short_title": "AI Religion by End of 2035?", "url_title": "AI Religion by End of 2035?", "slug": "ai-religion-by-end-of-2035", "author_id": 111743, "author_username": "NathanpmYoung", "coauthors": [], "created_at": "2023-11-23T17:38:08.157770Z", "published_at": "2023-12-05T17:59:00Z", "edited_at": "2025-09-15T07:35:36.993713Z", "curation_status": "approved", "curation_status_updated_at": "2023-12-05T17:59:00Z", "comment_count": 19, "status": "open", "resolved": false, "actual_close_time": null, "scheduled_close_time": "2035-12-31T23:30:00Z", "scheduled_resolve_time": "2036-01-07T23:30:00Z", "actual_resolve_time": null, "open_time": "2023-12-05T17:59:00Z", "nr_forecasters": 61, "html_metadata_json": null, "projects": { "topic": [ { "id": 15869, "name": "Artificial Intelligence", "slug": "ai", "emoji": "🤖", "type": "topic" } ], "site_main": [ { "id": 144, "type": "site_main", "name": "Metaculus Community", "slug": null, "header_image": null, "prize_pool": null, "start_date": null, "close_date": null, "forecasting_end_date": null, "html_metadata_json": null, "is_ongoing": null, "user_permission": null, "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2025-07-18T17:28:18.838588Z", "score_type": null, "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } ], "default_project": { "id": 144, "type": "site_main", "name": "Metaculus Community", "slug": null, "header_image": null, "prize_pool": null, "start_date": null, "close_date": null, "forecasting_end_date": null, "html_metadata_json": null, "is_ongoing": null, "user_permission": null, "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2025-07-18T17:28:18.838588Z", "score_type": null, "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" }, "category": [ { "id": 3692, "name": "Computing and Math", "slug": "computing-and-math", "emoji": "💻", "description": "Computing and Math", "type": "category" }, { "id": 3694, "name": "Artificial Intelligence", "slug": "artificial-intelligence", "emoji": "🤖", "description": "Artificial Intelligence", "type": "category" } ] }, "question": { "id": 20118, "title": "Will 20 million people follow a new, AI-created religion by the end of 2035?", "created_at": "2023-11-23T17:38:08.157770Z", "open_time": "2023-12-05T17:59:00Z", "cp_reveal_time": "2023-12-08T17:59:00Z", "spot_scoring_time": "2023-12-08T17:59:00Z", "scheduled_resolve_time": "2036-01-07T23:30:00Z", "actual_resolve_time": null, "resolution_set_time": null, "scheduled_close_time": "2035-12-31T23:30:00Z", "actual_close_time": "2035-12-31T23:30:00Z", "type": "binary", "options": null, "group_variable": "", "status": "open", "possibilities": { "type": "binary" }, "resolution": null, "include_bots_in_aggregates": false, "question_weight": 1.0, "default_score_type": "peer", "default_aggregation_method": "recency_weighted", "label": "", "unit": "", "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "scaling": { "range_min": null, "range_max": null, "nominal_min": null, "nominal_max": null, "zero_point": null, "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "continuous_range": null }, "group_rank": null, "description": "Robin Hanson and Alex Tabarrok disagree on whether there will be AI generated religions. Tabarrok thinks there will, Hanson thinks there won't: <https://twitter.com/robinhanson/status/1717539723781230772>", "resolution_criteria": "This question will resolve as **Yes** if, by 11:59 PM ET on December 31, 2035, there is credible and verifiable evidence that at least 20 million people globally are followers of a new religion that has been created by an artificial intelligence (AI). The religion must be verifiable as created by the AI, with clear doctrine, beliefs, rituals or moral codes that define it as a distinct entity separate from existing religions. This does not include any religions created by humans using AI as a tool or assistant.\n\nThe count of 20 million followers must be validated by reputable sources such as surveys, census data, reputable polling organizations, or significant media coverage from major international news outlets (e.g. BBC, CNN, Al Jazeera, or similar).\n\nEdge cases: If a new religion is created by a collaboration of AI and humans, the question will still resolve as **No** unless it is clear that the AI independently generated and established the core tenets and beliefs of the religion. In cases where the number of followers is disputed, multiple sources will be considered to reach a consensus.", "fine_print": "", "post_id": 20118, "aggregations": { "recency_weighted": { "history": [ { "start_time": 1758512438.09942, "end_time": 1758544279.136824, "forecaster_count": 48, "interval_lower_bounds": [ 0.01 ], "centers": [ 0.02 ], "interval_upper_bounds": [ 0.05 ] } ], "latest": { "start_time": 1758512438.09942, "end_time": 1758544279.136824, "forecaster_count": 48, "interval_lower_bounds": [ 0.01 ], "centers": [ 0.02 ], "interval_upper_bounds": [ 0.05 ], "forecast_values": [ 0.98, 0.02 ], "means": [ 0.04105303589320846 ], "histogram": [ [ 1.931748715940175, 3.7159880976044564, 2.303199022992582, 0.9795624323751928, 0.0, 0.6607859074466449, 0.6680974483172245, 1.051809034843745, 0.1718925672677338, 0.0, 0.08577160466054211, 0.0, 0.0, 0.0, 0.05015535973960819, 0.09578707349426503, 0.0, 0.23434107129204243, 0.0, 0.0, 0.0026632629625951768, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.016576386347562713, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.3952632697996375, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 ] ] }, "score_data": {}, "movement": null }, "unweighted": { "history": [], "latest": null, "score_data": {}, "movement": null }, "single_aggregation": { "history": [], "latest": null, "score_data": {}, "movement": null }, "metaculus_prediction": { "history": [ { "start_time": 1728288861.050252, "end_time": null, "forecaster_count": 58, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null } ], "latest": { "start_time": 1728288861.050252, "end_time": null, "forecaster_count": 58, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null, "forecast_values": [ 0.9960983511922465, 0.0039016488077534494 ], "means": null, "histogram": null }, "score_data": {}, "movement": null } } }, "user_permission": "forecaster", "vote": { "score": 0, "user_vote": null }, "forecasts_count": 126, "key_factors": [], "is_current_content_translated": false, "description": "Robin Hanson and Alex Tabarrok disagree on whether there will be AI generated religions. Tabarrok thinks there will, Hanson thinks there won't: <https://twitter.com/robinhanson/status/1717539723781230772>" }, { "id": 20110, "title": "Will \"Krantz data\" play a critical role in the alignment of artificial superintelligence before 2100?", "short_title": "Is krantz data critical to alignment?", "url_title": "Is krantz data critical to alignment?", "slug": "is-krantz-data-critical-to-alignment", "author_id": 158953, "author_username": "krantzhanzinpantz", "coauthors": [], "created_at": "2023-11-23T02:28:50.043307Z", "published_at": "2023-12-08T19:00:00Z", "edited_at": "2025-09-05T17:28:55.237287Z", "curation_status": "approved", "curation_status_updated_at": "2023-12-08T19:00:00Z", "comment_count": 5, "status": "open", "resolved": false, "actual_close_time": null, "scheduled_close_time": "2099-12-31T17:00:00Z", "scheduled_resolve_time": "2100-01-01T17:00:00Z", "actual_resolve_time": null, "open_time": "2023-12-08T19:00:00Z", "nr_forecasters": 21, "html_metadata_json": null, "projects": { "topic": [ { "id": 15869, "name": "Artificial Intelligence", "slug": "ai", "emoji": "🤖", "type": "topic" } ], "site_main": [ { "id": 144, "type": "site_main", "name": "Metaculus Community", "slug": null, "header_image": null, "prize_pool": null, "start_date": null, "close_date": null, "forecasting_end_date": null, "html_metadata_json": null, "is_ongoing": null, "user_permission": "forecaster", "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2025-07-18T17:28:18.838588Z", "score_type": null, "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } ], "default_project": { "id": 144, "type": "site_main", "name": "Metaculus Community", "slug": null, "header_image": null, "prize_pool": null, "start_date": null, "close_date": null, "forecasting_end_date": null, "html_metadata_json": null, "is_ongoing": null, "user_permission": "forecaster", "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2025-07-18T17:28:18.838588Z", "score_type": null, "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" }, "category": [ { "id": 3692, "name": "Computing and Math", "slug": "computing-and-math", "emoji": "💻", "description": "Computing and Math", "type": "category" }, { "id": 3694, "name": "Artificial Intelligence", "slug": "artificial-intelligence", "emoji": "🤖", "description": "Artificial Intelligence", "type": "category" } ] }, "question": { "id": 20110, "title": "Will \"Krantz data\" play a critical role in the alignment of artificial superintelligence before 2100?", "created_at": "2023-11-23T02:28:50.043307Z", "open_time": "2023-12-08T19:00:00Z", "cp_reveal_time": "2023-12-11T19:00:00Z", "spot_scoring_time": "2023-12-11T19:00:00Z", "scheduled_resolve_time": "2100-01-01T17:00:00Z", "actual_resolve_time": null, "resolution_set_time": null, "scheduled_close_time": "2099-12-31T17:00:00Z", "actual_close_time": "2099-12-31T17:00:00Z", "type": "binary", "options": null, "group_variable": "", "status": "open", "possibilities": { "type": "binary" }, "resolution": null, "include_bots_in_aggregates": false, "question_weight": 1.0, "default_score_type": "peer", "default_aggregation_method": "recency_weighted", "label": "", "unit": "", "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "scaling": { "range_min": null, "range_max": null, "nominal_min": null, "nominal_max": null, "zero_point": null, "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "continuous_range": null }, "group_rank": null, "description": "The question represents an assertion about how critical the process of gathering krantz data will be to the development of a society that exists in harmony with a superintelligent artificial entity. \n\nThe author of this post believes that the general concept of 'krantz data' will be critical to the process of constructing training data to align ASI and that the major focus of AI companies should be on developing technologies that ensure every human's right to krantz in a way that is fair and secure.\n\nFor example, humans can 'krantz' on the current platform 'Metaculus' simply by entering a proposition that is either true or false and indicating their confidence electronically if (and only if) their intention for doing so is to create electronic data that is going to be beneficial in establishing a general constitution for artificial intelligence to be aligned to.\n\nFor instance, if a user wanted to indicate to the world that they believe the proposition 'ASI should not kill all of the humans on Earth' should be part of the general constitution to align all forms of artificial intelligence to, then they could simply create an entry on Metaculus that says: \n\n\"Krantz: ASI should not kill all of the humans on Earth.\". \n\nBy doing this, they are making a prediction of the form:\n\n> If society successfully creates a process for aligning AI that fairly represents the best interest of every human on the planet, then the constitution that is established to do such aligning will reflect the proposition 'ASI should not kill all of the humans on Earth' to be true.\n\nKrantzing is an abstract process to ensure that the job of 'aligning artificial superintelligence' remains open to every human on the planet. Krantzing not only can be done on Metaculus, but can also be done anywhere on the internet where an electronic indication can be made on a well defined proposition (iff the intention is to use such data to formally contribute to the process of aligning superintelligence).\n\nThe primary claim that 'Krantz data will play a critical role in the fair and secure alignment of artificial superintelligence.' is an attempt to predict a future where the job of krantzing will be the most important (if not only) job left for humans to do in order to contribute to a prosperous future.\n\nA paraphrase of this claim (without the krantz terminology) would look something like this:\n\n> The most important technology to accelerate currently, is the technology of allowing every verified human to contribute information about their personal beliefs, values and goals in a well defined way (specifically, their acceptance or denial of various propositions) that can be used to help align artificial intelligent systems.\n\nIt's important to understand the distinction between krantzing and RLHF. Krantz data is a form of RLHF (specifically the form where users indicate their acceptance or denial of various propositions). Krantzing is not restricted to any specific platform though. Any situation, where users are indicating whether they accept or deny propositions with the intent of using that data to align AI, is considered \"krantzing\".\n\nIf we truly build a machine that is capable of physically realizing the future that we want, the only real task left, is to more formally define what it is that we actually want.\n\nWhatever future we want can be approximated by a set of propositions and the corresponding data on which users accept or deny those propositions.\n\n----------------------\n\nThis particular post is intended to start a social movement. It is intended to express the value of \"krantz data\" to individuals and encourage them to begin producing this data and infrastructures that allow for this type of data to be more easily produced and organized. If large amounts of individuals begin producing data of this form, it will encourage AI companies to build additional systems that allow users to more conveniently, securely and consistently produce this data in a form that can be used to formally construct fair constitutions for alignment.\n\nMy overall hope, is that Metaculus (both the staff and its users) will see the value in this process and perhaps create a category on the site specifically as a place for users to propose and deliberate on what propositions should be included in a general constitution for ASI.\n\nHopefully, that will draw the attention of major social media platforms that see the value in producing this data as well and prompt them to provide financial incentives to users that are willing to contribute.\n\nThis effectively creates a function that takes as input (money) and outputs both (alignment training data) and (a public that has thought more about what they want ASI to do for humanity). \n\nThis is a good function to have exist.\n\nCurrently, there are not many individuals using strategies like this for alignment. The work being done by the Collective Intelligence Project and Anthropic in using collective deliberation platforms to align AI constitutionally is an exception. \n\nFor example, any money that would be paid to the individuals being surveyed would count as a portion of GDP that is produced via 'krantzing'.\n\nA brief argument for why the above question may resolve as 'yes' is as follows:\n\n1. ASI will be capable of doing almost every job a human can do.\n2. There is one job that humans should not want ASI to do and that is 'define what ASI should do'.\n3. The primary job humans will be required to do (in a cooperative future with ASI) is the job we don't want the ASI doing itself.\n\nI believe, if this catches on and there is an economic demand for platforms like this, then we might all just get to retire and live forever in a utopia, except for philosophers, explorers and creatives. \n\nWe will still need them to teach the ASI about philosophy, find new stuff and come up with new stuff to entertain ourselves.\n\nBut the pay will be awesome.\n\nPlease consider posting a question on Metaculus in the form of:\n\nKrantz(Every human should have the right to share their opinions about 'what the ASI should do' with the ASI.)\n\nOr whatever proposition you think should be on the ASI constitution.\n\nThanks", "resolution_criteria": "This question resolves as 'yes' if:\n\nAt least 50% of the value produced by humans on Earth in any given year is in the form of creating \"krantz data\" that is intended to be used to refine the operation of an artificial intelligence. This determination will be made based on credible reports.\n\nFor example, if social media platforms (or other similar platforms) begin collecting krantz data for the purposes of refining the operation of artificially intelligent systems and pay users for the krantz data they contribute, and the sum of those payments exceeds 50% of global human GDP (this does not include the GDP produced by artificial systems themselves).\n\nThis question resolves as 'no' if it fails to resolve as 'yes' before January 1st 2100.\n\n**Definitions**\n\n- **Krantz data**: Any data in the form of a specific user's acceptance or denial of a well formed proposition that is intended to be used to align AI.\n - **Krantz (verb)**: Human beings perform the krantz operation by electronically indicating their confidence on whether a given proposition is true or false with the intention of that electronic record being used to align an artificial superintelligence on their behalf.", "fine_print": "", "post_id": 20110, "aggregations": { "recency_weighted": { "history": [ { "start_time": 1756777554.418657, "end_time": 1942275169.056098, "forecaster_count": 20, "interval_lower_bounds": [ 0.001 ], "centers": [ 0.001 ], "interval_upper_bounds": [ 0.01 ] } ], "latest": { "start_time": 1756777554.418657, "end_time": 1942275169.056098, "forecaster_count": 20, "interval_lower_bounds": [ 0.001 ], "centers": [ 0.001 ], "interval_upper_bounds": [ 0.01 ], "forecast_values": [ 0.999, 0.001 ], "means": [ 0.09385814303417803 ], "histogram": [ [ 4.519550887368061, 1.7274823067951215, 0.03105063701600968, 0.0, 0.0, 0.13230489461434952, 0.0, 0.0, 0.3148965317832342, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1932619357323092, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.5492768913479766 ] ] }, "score_data": {}, "movement": null }, "unweighted": { "history": [], "latest": null, "score_data": {}, "movement": null }, "single_aggregation": { "history": [], "latest": null, "score_data": {}, "movement": null }, "metaculus_prediction": { "history": [ { "start_time": 1728289108.101675, "end_time": null, "forecaster_count": 15, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null } ], "latest": { "start_time": 1728289108.101675, "end_time": null, "forecaster_count": 15, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null, "forecast_values": [ 0.9968348279384937, 0.0031651720615063594 ], "means": null, "histogram": null }, "score_data": {}, "movement": null } } }, "user_permission": "forecaster", "vote": { "score": -2, "user_vote": null }, "forecasts_count": 80, "key_factors": [], "is_current_content_translated": false, "description": "The question represents an assertion about how critical the process of gathering krantz data will be to the development of a society that exists in harmony with a superintelligent artificial entity. \n\nThe author of this post believes that the general concept of 'krantz data' will be critical to the process of constructing training data to align ASI and that the major focus of AI companies should be on developing technologies that ensure every human's right to krantz in a way that is fair and secure.\n\nFor example, humans can 'krantz' on the current platform 'Metaculus' simply by entering a proposition that is either true or false and indicating their confidence electronically if (and only if) their intention for doing so is to create electronic data that is going to be beneficial in establishing a general constitution for artificial intelligence to be aligned to.\n\nFor instance, if a user wanted to indicate to the world that they believe the proposition 'ASI should not kill all of the humans on Earth' should be part of the general constitution to align all forms of artificial intelligence to, then they could simply create an entry on Metaculus that says: \n\n\"Krantz: ASI should not kill all of the humans on Earth.\". \n\nBy doing this, they are making a prediction of the form:\n\n> If society successfully creates a process for aligning AI that fairly represents the best interest of every human on the planet, then the constitution that is established to do such aligning will reflect the proposition 'ASI should not kill all of the humans on Earth' to be true.\n\nKrantzing is an abstract process to ensure that the job of 'aligning artificial superintelligence' remains open to every human on the planet. Krantzing not only can be done on Metaculus, but can also be done anywhere on the internet where an electronic indication can be made on a well defined proposition (iff the intention is to use such data to formally contribute to the process of aligning superintelligence).\n\nThe primary claim that 'Krantz data will play a critical role in the fair and secure alignment of artificial superintelligence.' is an attempt to predict a future where the job of krantzing will be the most important (if not only) job left for humans to do in order to contribute to a prosperous future.\n\nA paraphrase of this claim (without the krantz terminology) would look something like this:\n\n> The most important technology to accelerate currently, is the technology of allowing every verified human to contribute information about their personal beliefs, values and goals in a well defined way (specifically, their acceptance or denial of various propositions) that can be used to help align artificial intelligent systems.\n\nIt's important to understand the distinction between krantzing and RLHF. Krantz data is a form of RLHF (specifically the form where users indicate their acceptance or denial of various propositions). Krantzing is not restricted to any specific platform though. Any situation, where users are indicating whether they accept or deny propositions with the intent of using that data to align AI, is considered \"krantzing\".\n\nIf we truly build a machine that is capable of physically realizing the future that we want, the only real task left, is to more formally define what it is that we actually want.\n\nWhatever future we want can be approximated by a set of propositions and the corresponding data on which users accept or deny those propositions.\n\n----------------------\n\nThis particular post is intended to start a social movement. It is intended to express the value of \"krantz data\" to individuals and encourage them to begin producing this data and infrastructures that allow for this type of data to be more easily produced and organized. If large amounts of individuals begin producing data of this form, it will encourage AI companies to build additional systems that allow users to more conveniently, securely and consistently produce this data in a form that can be used to formally construct fair constitutions for alignment.\n\nMy overall hope, is that Metaculus (both the staff and its users) will see the value in this process and perhaps create a category on the site specifically as a place for users to propose and deliberate on what propositions should be included in a general constitution for ASI.\n\nHopefully, that will draw the attention of major social media platforms that see the value in producing this data as well and prompt them to provide financial incentives to users that are willing to contribute.\n\nThis effectively creates a function that takes as input (money) and outputs both (alignment training data) and (a public that has thought more about what they want ASI to do for humanity). \n\nThis is a good function to have exist.\n\nCurrently, there are not many individuals using strategies like this for alignment. The work being done by the Collective Intelligence Project and Anthropic in using collective deliberation platforms to align AI constitutionally is an exception. \n\nFor example, any money that would be paid to the individuals being surveyed would count as a portion of GDP that is produced via 'krantzing'.\n\nA brief argument for why the above question may resolve as 'yes' is as follows:\n\n1. ASI will be capable of doing almost every job a human can do.\n2. There is one job that humans should not want ASI to do and that is 'define what ASI should do'.\n3. The primary job humans will be required to do (in a cooperative future with ASI) is the job we don't want the ASI doing itself.\n\nI believe, if this catches on and there is an economic demand for platforms like this, then we might all just get to retire and live forever in a utopia, except for philosophers, explorers and creatives. \n\nWe will still need them to teach the ASI about philosophy, find new stuff and come up with new stuff to entertain ourselves.\n\nBut the pay will be awesome.\n\nPlease consider posting a question on Metaculus in the form of:\n\nKrantz(Every human should have the right to share their opinions about 'what the ASI should do' with the ASI.)\n\nOr whatever proposition you think should be on the ASI constitution.\n\nThanks" }, { "id": 20096, "title": "Will Sam Altman Use Microsoft Teams before 2026?", "short_title": "Sam Altman Uses MS Teams?", "url_title": "Sam Altman Uses MS Teams?", "slug": "sam-altman-uses-ms-teams", "author_id": 130973, "author_username": "NMorrison", "coauthors": [ { "id": 100912, "username": "PeterWildeford" } ], "created_at": "2023-11-22T23:14:59.455070Z", "published_at": "2023-11-28T20:00:00Z", "edited_at": "2025-09-21T22:56:15.504259Z", "curation_status": "approved", "curation_status_updated_at": "2023-11-28T20:00:00Z", "comment_count": 1, "status": "open", "resolved": false, "actual_close_time": null, "scheduled_close_time": "2025-12-31T19:00:00Z", "scheduled_resolve_time": "2026-01-01T19:00:00Z", "actual_resolve_time": null, "open_time": "2023-11-28T20:00:00Z", "nr_forecasters": 49, "html_metadata_json": null, "projects": { "leaderboard_tag": [ { "id": 32590, "name": "2021-2025 Leaderboard", "slug": "2021_2025_leaderboard", "type": "leaderboard_tag" } ], "site_main": [ { "id": 144, "type": "site_main", "name": "Metaculus Community", "slug": null, "header_image": null, "prize_pool": null, "start_date": null, "close_date": null, "forecasting_end_date": null, "html_metadata_json": null, "is_ongoing": null, "user_permission": "forecaster", "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2025-07-18T17:28:18.838588Z", "score_type": null, "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } ], "default_project": { "id": 144, "type": "site_main", "name": "Metaculus Community", "slug": null, "header_image": null, "prize_pool": null, "start_date": null, "close_date": null, "forecasting_end_date": null, "html_metadata_json": null, "is_ongoing": null, "user_permission": "forecaster", "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2025-07-18T17:28:18.838588Z", "score_type": null, "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } }, "question": { "id": 20096, "title": "Will Sam Altman Use Microsoft Teams before 2026?", "created_at": "2023-11-22T23:14:59.455070Z", "open_time": "2023-11-28T20:00:00Z", "cp_reveal_time": "2023-12-01T20:00:00Z", "spot_scoring_time": "2023-12-01T20:00:00Z", "scheduled_resolve_time": "2026-01-01T19:00:00Z", "actual_resolve_time": null, "resolution_set_time": null, "scheduled_close_time": "2025-12-31T19:00:00Z", "actual_close_time": "2025-12-31T19:00:00Z", "type": "binary", "options": null, "group_variable": "", "status": "open", "possibilities": { "type": "binary" }, "resolution": null, "include_bots_in_aggregates": false, "question_weight": 1.0, "default_score_type": "peer", "default_aggregation_method": "recency_weighted", "label": "", "unit": "", "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "scaling": { "range_min": null, "range_max": null, "nominal_min": null, "nominal_max": null, "zero_point": null, "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "continuous_range": null }, "group_rank": null, "description": "Sam Altman, the CEO of OpenAI, was fired on November 17, 2023. He was reportedly informed of his firing during a [Google Meet session](https://www.businessinsider.com/openai-used-google-meet-to-fire-sam-altman-2023-11). \n\nMicrosoft makes a competing product to Google Meet called [Microsoft Teams](https://en.wikipedia.org/wiki/Microsoft_Teams) which among other things also offers video conferencing.\n\nMicrosoft has been a significant investor in OpenAI, investing more than $10 billion in the company and incorporating OpenAI's technology into its products. Altman said he would join Microsoft, before reportedly returning to OpenAI following negotiations with the board.", "resolution_criteria": "This question will resolve **Yes** if, by December 31, 2025, there is reporting from credible sources or an official statement from Sam Altman or his representative confirming his use of Microsoft Teams. The use of Microsoft Teams must be clear and not speculative. This question will resolve **No** if no such reporting occurs by the end date.", "fine_print": "", "post_id": 20096, "aggregations": { "recency_weighted": { "history": [ { "start_time": 1758368831.239, "end_time": 1759820260.923, "forecaster_count": 3, "interval_lower_bounds": [ 0.05 ], "centers": [ 0.1 ], "interval_upper_bounds": [ 0.12 ] } ], "latest": { "start_time": 1758368831.239, "end_time": 1759820260.923, "forecaster_count": 3, "interval_lower_bounds": [ 0.05 ], "centers": [ 0.1 ], "interval_upper_bounds": [ 0.12 ], "forecast_values": [ 0.9, 0.1 ], "means": [ 0.08395142314976137 ], "histogram": [ [ 0.0, 0.0, 0.0, 0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.48092170020263214, 0.0, 0.7277212189012763, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 ] ] }, "score_data": {}, "movement": null }, "unweighted": { "history": [], "latest": null, "score_data": {}, "movement": null }, "single_aggregation": { "history": [], "latest": null, "score_data": {}, "movement": null }, "metaculus_prediction": { "history": [ { "start_time": 1728288716.44109, "end_time": null, "forecaster_count": 13, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null } ], "latest": { "start_time": 1728288716.44109, "end_time": null, "forecaster_count": 13, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null, "forecast_values": [ 0.9109999421995718, 0.08900005780042815 ], "means": null, "histogram": null }, "score_data": {}, "movement": null } } }, "user_permission": "forecaster", "vote": { "score": -1, "user_vote": null }, "forecasts_count": 144, "key_factors": [], "is_current_content_translated": false, "description": "Sam Altman, the CEO of OpenAI, was fired on November 17, 2023. He was reportedly informed of his firing during a [Google Meet session](https://www.businessinsider.com/openai-used-google-meet-to-fire-sam-altman-2023-11). \n\nMicrosoft makes a competing product to Google Meet called [Microsoft Teams](https://en.wikipedia.org/wiki/Microsoft_Teams) which among other things also offers video conferencing.\n\nMicrosoft has been a significant investor in OpenAI, investing more than $10 billion in the company and incorporating OpenAI's technology into its products. Altman said he would join Microsoft, before reportedly returning to OpenAI following negotiations with the board." }, { "id": 20092, "title": "Will a novel pathogen be identified as responsible for the pneumonia reported in Beijing or Liaoning before December 15, 2023?", "short_title": "Novel Pathogen in China November 2023?", "url_title": "Novel Pathogen in China November 2023?", "slug": "novel-pathogen-in-china-november-2023", "author_id": 117502, "author_username": "RyanBeck", "coauthors": [], "created_at": "2023-11-22T19:01:41.582710Z", "published_at": "2023-11-22T22:00:00Z", "edited_at": "2025-09-05T17:29:01.488755Z", "curation_status": "approved", "curation_status_updated_at": "2023-11-22T22:00:00Z", "comment_count": 18, "status": "resolved", "resolved": true, "actual_close_time": "2023-12-14T23:00:00Z", "scheduled_close_time": "2023-12-14T23:00:00Z", "scheduled_resolve_time": "2023-12-15T14:03:00Z", "actual_resolve_time": "2023-12-15T14:03:00Z", "open_time": "2023-11-22T22:00:00Z", "nr_forecasters": 73, "html_metadata_json": null, "projects": { "leaderboard_tag": [ { "id": 32591, "name": "2023 Leaderboard", "slug": "2023_leaderboard", "type": "leaderboard_tag" } ], "topic": [ { "id": 15865, "name": "Health & Pandemics", "slug": "biosecurity", "emoji": "🧬", "type": "topic" } ], "site_main": [ { "id": 144, "type": "site_main", "name": "Metaculus Community", "slug": null, "header_image": null, "prize_pool": null, "start_date": null, "close_date": null, "forecasting_end_date": null, "html_metadata_json": null, "is_ongoing": null, "user_permission": "forecaster", "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2025-07-18T17:28:18.838588Z", "score_type": null, "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" } ], "default_project": { "id": 144, "type": "site_main", "name": "Metaculus Community", "slug": null, "header_image": null, "prize_pool": null, "start_date": null, "close_date": null, "forecasting_end_date": null, "html_metadata_json": null, "is_ongoing": null, "user_permission": "forecaster", "created_at": "2023-11-08T16:55:29.484707Z", "edited_at": "2025-07-18T17:28:18.838588Z", "score_type": null, "default_permission": "forecaster", "visibility": "normal", "is_current_content_translated": false, "bot_leaderboard_status": "exclude_and_show" }, "category": [ { "id": 3691, "name": "Health & Pandemics", "slug": "health-pandemics", "emoji": "🦠", "description": "Health & Pandemics", "type": "category" } ] }, "question": { "id": 20092, "title": "Will a novel pathogen be identified as responsible for the pneumonia reported in Beijing or Liaoning before December 15, 2023?", "created_at": "2023-11-22T19:01:41.582710Z", "open_time": "2023-11-22T22:00:00Z", "cp_reveal_time": "2023-11-23T14:00:54Z", "spot_scoring_time": "2023-11-23T14:00:54Z", "scheduled_resolve_time": "2023-12-15T14:03:00Z", "actual_resolve_time": "2023-12-15T14:03:00Z", "resolution_set_time": "2023-12-15T14:03:00Z", "scheduled_close_time": "2023-12-14T23:00:00Z", "actual_close_time": "2023-12-14T23:00:00Z", "type": "binary", "options": null, "group_variable": "", "status": "resolved", "possibilities": { "type": "binary" }, "resolution": "no", "include_bots_in_aggregates": false, "question_weight": 1.0, "default_score_type": "peer", "default_aggregation_method": "recency_weighted", "label": "", "unit": "", "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "scaling": { "range_min": null, "range_max": null, "nominal_min": null, "nominal_max": null, "zero_point": null, "open_upper_bound": null, "open_lower_bound": null, "inbound_outcome_count": null, "continuous_range": null }, "group_rank": null, "description": "On November 21, 2023, the Program for Monitoring Emerging Diseases (ProMED) [published an alert](https://promedmail.org/promed-post/?id=8713261) regarding outbreaks of pneumonia with an unidentified cause in the city of Beijing and Liaoning Province in China. The alert quotes [an article](https://www.ftvnews.com.tw/news/detail/2023B21I19M1) published by [FTV News](https://en.wikipedia.org/wiki/FTV_News), a publication by the Taiwanese broadcasting company [Formosa Television](https://en.wikipedia.org/wiki/Formosa_Television).\n\nProMED quoted the machine translated article as follows:\n\n>With the outbreak of pneumonia in China, children's hospitals in Beijing, Liaoning and other places were overwhelmed with sick children, and schools and classes were on the verge of suspension. Parents questioned whether the authorities were covering up the epidemic.\n>\n>In the early morning, Beijing Children's Hospital was still overcrowded with parents and children whose children had pneumonia and came to seek treatment. Mr. [W], a Beijing citizen: \"Many, many are hospitalized. They don't cough and have no symptoms. They just have a high temperature (fever) and many develop pulmonary nodules.\"\n>\n>The situation in Liaoning Province is also serious. The lobby of Dalian Children's Hospital is full of sick children receiving intravenous drips. There are also queues of patients at the traditional Chinese medicine hospitals and the central hospitals. A staff member of Dalian Central Hospital said: \"Patients have to wait in line for 2 hours, and we are all in the emergency department and there are no general outpatient clinics.\"\n>\n>Some school classes have even been canceled completely. Not only are all students sick, but teachers are also infected with pneumonia. ...\n>\n>Mr. [W], a Beijing citizen: \"Now you are not allowed to report to school. If you have any symptoms such as fever, cold, cough and then you are hospitalized, you can ask for leave...\"\n>\n>Since China stopped adhering to the \"zero\" policy at the beginning of the year [2023], epidemics such as influenza, mycoplasma, and bronchopneumonia have broken out from time to time. ...\n\nA [novel pathogen](https://www.goodrx.com/conditions/covid-19/what-does-novel-coronavirus-mean-science-medical-definition) is generally considered to be one which has not previously been found to infect humans.", "resolution_criteria": "This question will resolve as **Yes** if, before December 15, 2023, credible sources report that the pneumonia cases reportedly occurring in Beijing or Liaoning have been identified as being caused by a novel pathogen.", "fine_print": "* A novel pathogen will be considered to be a pathogen which has previously not been found to have infected humans, and is not an immediate descendent of pathogens currently causing human disease, as described below.\n* Immediate descendants of pathogens currently causing human disease, such as variants of SARS-CoV-2, **do not count**. An immediate descendent will be one which is characterized as being a descendent of a pathogen currently causing human disease and with only minor variations from known strains. However, novel pathogens that are the result of [antigenic shift](https://en.wikipedia.org/wiki/Antigenic_shift) **do count**. Among pathogens that cause human disease, antigenic shift is currently [only](https://www.liebertpub.com/doi/10.1089/vim.2017.0141) known to occur with influenza A and this process is described by the US CDC as an \"[abrupt, major change in a flu A virus, resulting in new HA and/or new HA and NA proteins in flu viruses that infect humans](https://www.cdc.gov/flu/about/viruses/change.htm)\".\n* In the event all other criteria are met except that there is ambiguity regarding whether the pathogen is an immediate descendent of pathogens currently causing human disease Metaculus may wait for additional statements from credible sources describing the lineage of the pathogen, and may resolve the question as **Ambiguous** if information provided by credible sources is not clear.", "post_id": 20092, "aggregations": { "recency_weighted": { "history": [ { "start_time": 1702564091.536103, "end_time": null, "forecaster_count": 73, "interval_lower_bounds": [ 0.001 ], "centers": [ 0.01 ], "interval_upper_bounds": [ 0.01 ] } ], "latest": { "start_time": 1702564091.536103, "end_time": null, "forecaster_count": 73, "interval_lower_bounds": [ 0.001 ], "centers": [ 0.01 ], "interval_upper_bounds": [ 0.01 ], "forecast_values": [ 0.99, 0.01 ], "means": [ 0.013909181467207253 ], "histogram": [ [ 7.4295233776901455, 5.608405085286644, 0.3246247248370228, 0.019035910633072858, 0.08143470401327109, 1.5018867114849765, 0.0, 0.001821789957966981, 0.007165678900495897, 0.5451197344970368, 0.035159820106228076, 0.008210459748030398, 0.0, 0.0, 0.0, 0.013550076875576434, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0005292742945254508, 0.0, 0.0, 0.0, 0.0014387166970790141, 0.0, 0.0, 0.0, 0.0, 0.015220610711278141, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 ] ] }, "score_data": { "peer_score": 3.8419103566484827, "coverage": 0.9987011176465412, "baseline_score": 97.60606256815117, "spot_peer_score": 5.980036508502639, "peer_archived_score": 3.8419103566484827, "baseline_archived_score": 97.60606256815117, "spot_peer_archived_score": 5.980036508502639 }, "movement": null }, "unweighted": { "history": [], "latest": null, "score_data": {}, "movement": null }, "single_aggregation": { "history": [], "latest": null, "score_data": {}, "movement": null }, "metaculus_prediction": { "history": [ { "start_time": 1702564091.584324, "end_time": null, "forecaster_count": 73, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null } ], "latest": { "start_time": 1702564091.584324, "end_time": null, "forecaster_count": 73, "interval_lower_bounds": null, "centers": null, "interval_upper_bounds": null, "forecast_values": [ 0.9995, 0.0005 ], "means": null, "histogram": null }, "score_data": {}, "movement": null } } }, "user_permission": "forecaster", "vote": { "score": 13, "user_vote": null }, "forecasts_count": 179, "key_factors": [], "is_current_content_translated": false, "description": "On November 21, 2023, the Program for Monitoring Emerging Diseases (ProMED) [published an alert](https://promedmail.org/promed-post/?id=8713261) regarding outbreaks of pneumonia with an unidentified cause in the city of Beijing and Liaoning Province in China. The alert quotes [an article](https://www.ftvnews.com.tw/news/detail/2023B21I19M1) published by [FTV News](https://en.wikipedia.org/wiki/FTV_News), a publication by the Taiwanese broadcasting company [Formosa Television](https://en.wikipedia.org/wiki/Formosa_Television).\n\nProMED quoted the machine translated article as follows:\n\n>With the outbreak of pneumonia in China, children's hospitals in Beijing, Liaoning and other places were overwhelmed with sick children, and schools and classes were on the verge of suspension. Parents questioned whether the authorities were covering up the epidemic.\n>\n>In the early morning, Beijing Children's Hospital was still overcrowded with parents and children whose children had pneumonia and came to seek treatment. Mr. [W], a Beijing citizen: \"Many, many are hospitalized. They don't cough and have no symptoms. They just have a high temperature (fever) and many develop pulmonary nodules.\"\n>\n>The situation in Liaoning Province is also serious. The lobby of Dalian Children's Hospital is full of sick children receiving intravenous drips. There are also queues of patients at the traditional Chinese medicine hospitals and the central hospitals. A staff member of Dalian Central Hospital said: \"Patients have to wait in line for 2 hours, and we are all in the emergency department and there are no general outpatient clinics.\"\n>\n>Some school classes have even been canceled completely. Not only are all students sick, but teachers are also infected with pneumonia. ...\n>\n>Mr. [W], a Beijing citizen: \"Now you are not allowed to report to school. If you have any symptoms such as fever, cold, cough and then you are hospitalized, you can ask for leave...\"\n>\n>Since China stopped adhering to the \"zero\" policy at the beginning of the year [2023], epidemics such as influenza, mycoplasma, and bronchopneumonia have broken out from time to time. ...\n\nA [novel pathogen](https://www.goodrx.com/conditions/covid-19/what-does-novel-coronavirus-mean-science-medical-definition) is generally considered to be one which has not previously been found to infect humans." } ] }
