Question Feed | Metaculus

Topics

🗳️

US Democracy Index

❓

What will be the best performance on OSWorld by December 31st 2025?

Current estimate

73.1%

Contributed by the AI 2025 Forecasting Survey community.

1 comment

9 forecasters

What will be the best score on Cybench by December 31st 2025?

Current estimate

82.6%

26 comments

44 forecasters

Five years after AGI, will AI philosophical competence be solved?

11%chance

8 comments

31 forecasters

When will a computer program be listed as an author in journal article in the Annals of Mathematics?

Current estimate

01 Jun 2029

Contributed by the JetBrains community.

0 comments

3 forecasters

What percentage of problems from Long Code Arena do you estimate the top-performing AI model or agent will be able to solve by December'25?

Commit message generation99.5

Module summarization99.5

Library-based code generation97.3

Contributed by the Harvard's AI Safety Student Team community.

0 comments

10 forecasters

What will be the best performance on SWE-bench Verified by December 31st 2025?

Latest estimate

94.2

This question is closed for forecasting.

Contributed by the AI 2025 Forecasting Survey community.

0 comments

15 forecasters

What will be the best performance on SWE-bench Verified by December 31st 2025?

Current estimate

87.5 score

condition

12 forecasters

20 comments

141 forecasters

When will an AI be able to solve over 80% of competition-level coding problems on its first try?

Current estimate

26 May 2027

Contributed by the AI 2025 Forecasting Survey community.

7 comments

21 forecasters

What will be the best performance on FrontierMath by December 31st 2025?

Current estimate

49.7%

What will be the best performance on OSWorld by December 31st 2025?

What will be the best score on Cybench by December 31st 2025?

Five years after AGI, will AI philosophical competence be solved?

When will a computer program be listed as an author in journal article in the Annals of Mathematics?

What percentage of problems from Long Code Arena do you estimate the top-performing AI model or agent will be able to solve by December'25?

What will be the best performance on SWE-bench Verified by December 31st 2025?

What will be the best performance on SWE-bench Verified by December 31st 2025?

US restricts compute before 2026?

Date AIs Capable of Developing AI Software

Date AIs Capable of Developing AI Software

When will an AI be able to solve over 80% of competition-level coding problems on its first try?

What will be the best performance on FrontierMath by December 31st 2025?

What will be the best performance on OSWorld by December 31st 2025?

What will be the best score on Cybench by December 31st 2025?

Five years after AGI, will AI philosophical competence be solved?

When will a computer program be listed as an author in journal article in the Annals of Mathematics?

What percentage of problems from Long Code Arena do you estimate the top-performing AI model or agent will be able to solve by December'25?

What will be the best performance on SWE-bench Verified by December 31st 2025?

What will be the best performance on SWE-bench Verified by December 31st 2025?

US restricts compute before 2026?

Date AIs Capable of Developing AI Software

Date AIs Capable of Developing AI Software

When will an AI be able to solve over 80% of competition-level coding problems on its first try?

What will be the best performance on FrontierMath by December 31st 2025?