JavaScript is disabled. Core navigation and editorial pages remain available, but live market widgets, charts, search, and personalization features may be limited.

Will AIs beat human experts in question-answering on the GPQA benchmark before January 1st, 2027?

One-OffAI Tech Science3mo

Manifold MarketsNo KYCVerified resolution dataFairly calibrated

Current community forecast

Yes 94.6%

Forecasters

32

Question type

binary

Methodology

Play-money forecasting platform

Source type

Forecast

Market data

Updated 10 hours ago

Stale

Jan 18, 24, 12:11 AM UTCJan 2, 27, 7:59 AM UTC

Trends

Outcome24hChance

Related Markets

Related questions for context, not a comparison of equivalent contracts.

In what year will AI achieve a score of 95% or higher on the GPQA benchmark?

Manifold Markets

Will AI beat top human players at Civ6 (without cheating) by EOY 2026?

Manifold Markets

Will the US and Iran be at war on January 1st, 2027?

Manifold Markets

AI wins IMO gold medal in 2026?

When will an AI lab solve another Millennium Prize Problem?

Which country will be the next to send humans to the Moon?

Rules

From the abstract,

We present GPQA, a challenging dataset of 448 multiple-choice questions written by domain experts in biology, physics, and chemistry.
We ensure that the questions are high-quality and extremely difficult: experts who have or are pursuing PhDs in the corresponding domains reach 65% accuracy (74% when discounting clear mistakes the experts identified in retrospect), while highly skilled non-expert validators only reach 34% accuracy, despite spending on average over 30 minutes with unrestricted access to the web (i.e., the questions are "Google-proof").
The questions are also difficult for state-of-the-art AI systems, with our strongest GPT-4 based baseline achieving 39% accuracy.
This question resolves to YES if a credible paper, blog post, or document of any kind indicates that at least some AI obtained a score of greater than 74.0% on the GPQA dataset before January 1st 2027, and NO otherwise.
The result must be credible, and I will exclude results that appear to be the result of cheating: for example, results obtained by training on the test set.

Open markets

Next Claude Opus released on?

September 22+43.5pt67.9%
September 24+1.0pt12.3%
September 23+0.9pt6.6%

+13 more outcomes

Total Vol: $251K
24h Vol: $205.2K

Time's Person of the Year for 2026

Zohran Mamdani-2.0pt27%
Mark Carney-1.0pt18%
Donald Trump10%

+39 more outcomes

Total Vol: $1.5M
24h Vol

Best AI at the end of 2026?

Claude-2.2pt69.9%
ChatGPT+0.8pt14.6%
Gemini-0.8pt7.8%

+5 more outcomes

Total Vol: $11.8M

Will Anthropic’s valuation hit __ by December 31?

↑$1.25T+1.3pt100%
↑$1.1T+3.6pt100%
↑$1.0T+2.4pt100%

+10 more outcomes

Which company has the best AI model end of September?

Anthropic+0.1pt98%
Google-0.3pt1%
OpenAI+0.3pt1%

+18 more outcomes

Which company has best AI model end of 2026?

Anthropic-1.0pt69.6%
OpenAI+3.5pt12.8%
Google-3.0pt9.8%

+12 more outcomes

Frequently asked questions

What are the current odds for Will AIs beat human experts in question-answering on the GPQA benchmark before January 1st, 2027??: Manifold Markets prices Yes at 95% as of September 21, 2026.
When does Will AIs beat human experts in question-answering on the GPQA benchmark before January 1st, 2027? resolve?: Will AIs beat human experts in question-answering on the GPQA benchmark before January 1st, 2027? is scheduled to resolve on January 2, 2027.
How is Will AIs beat human experts in question-answering on the GPQA benchmark before January 1st, 2027? settled?: Manifold Markets settles this market as follows: From the abstract,

Report a data issue with this market