Will an AI model outperform 95% of Manifold users on accuracy before 2026? | Manifold

Will an AI model outperform 95% of Manifold users on accuracy before 2026?

Plus

63

Ṁ3974

Dec 31

56%

chance

1D

1W

1M

ALL

This is comparing accuracy of the AI model vs Manifold users on a 'benchmark' selection of objective markets'.

Inspired by the Autocast competition: https://forecasting.mlsafety.org/

Close date updated to 2026-01-01 12:59 am

This question is managed and resolved by Manifold.

#Technical AI Timelines

Get

1,000

and

3.00

Sort by:

By users you mean human users right? Anything in the silicon league excluded?

Hmm... I think that this might depend crucially on how many inaccurate users Manifold will have at that point (e.g., would you count people who haven't been active for years, or all the people who lost their starting Mana in a few bets and then stopped using the platform)?

Not sure how you're measuring accuracy, but acc is more profitable and I don't have a better way to measure accuracy.

In accuracy or profit?

@Predictor accuracy. updated the market.

Related questions

Will Manifold stop using AI to make my questions worse by the end of 2025?

Will any AI music I release achieve a Manifold rating >= 5 in 2025?

Will there be an AI-powered trader on Manifold in the top 20 profit leaderboard by end of 2026?

Will a new lab create a top-performing AI frontier model before 2028?

What year will the first AI exceed 80% on MLE-bench?

Will an AI achieve >85% performance on the FrontierMath benchmark before 2028?

Will an AI achieve >85% performance on the FrontierMath benchmark before 2027?

An AI is trustworthy-ish on Manifold by 2030?

At the beginning of 2026, what percentage of Manifold users will believe that an AI intelligence explosion is a significant concern before 2075?

Will models be able to do the work of an AI researcher/engineer before 2027?

Related questions

Will Manifold stop using AI to make my questions worse by the end of 2025?

Will an AI achieve >85% performance on the FrontierMath benchmark before 2028?

Will any AI music I release achieve a Manifold rating >= 5 in 2025?

Will an AI achieve >85% performance on the FrontierMath benchmark before 2027?

Will there be an AI-powered trader on Manifold in the top 20 profit leaderboard by end of 2026?

An AI is trustworthy-ish on Manifold by 2030?

Will a new lab create a top-performing AI frontier model before 2028?

At the beginning of 2026, what percentage of Manifold users will believe that an AI intelligence explosion is a significant concern before 2075?

What year will the first AI exceed 80% on MLE-bench?

Will models be able to do the work of an AI researcher/engineer before 2027?

© Manifold Markets, Inc.•Terms + Mana-only Terms•Privacy•Rules