An option resolves YES if it is true about the AI model, or program, known to be State of the Art in terms of the FrontierMath benchmark, at the end of the year 2025. It resolves NO otherwise.

You're welcome to add any interesting facts that might or might not be true about the state of the art in math problems, as defined by achieving the highest score on the FrontierMath benchmarks.

I reserve the right to cancel any option that is too vague, too improbable, etc.

See also:
/Bayesian/what-will-true-of-the-sota-ai-on-th-y0LE5uE9n9 (This market)
/Bayesian/what-will-true-of-the-sota-ai-on-th-ROldIhZZgt
/Bayesian/what-will-true-of-the-sota-ai-on-th-RQptyR5uO8

/Bayesian/will-an-ai-achieve-85-performance-o-hyPtIE98qZ
/MatthewBarnett/will-an-ai-achieve-85-performance-o

/Bayesian/will-an-ai-achieve-30-performance-o

• Update 2025-11-06 (PST) (AI summary of creator comment): The creator is uncertain whether to use Epoch's independent evaluation (which has a weaker scaffold) or the highest reported score to determine the SOTA AI. This ambiguity in the resolution source has not yet been resolved.

• Update 2025-11-06 (PST) (AI summary of creator comment): If there is ambiguity about which model is truly SOTA (e.g., due to different evaluation methods or scaffolds), the creator will have someone uninvolved take a guess as to which model is truly state of the art, even if apple-to-apple comparisons are not possible.