09:00
2026-09-12
asiaai.fyi
ai-research
FrontierMath Benchmarks Saturated: Why Fields Medalists Are Resisting Brute-Force AI Models
OpenAI's GPT-6 Astra solved the final remaining problem in the elite FrontierMath Tier 4 benchmark, taking the model from a sub-2% success rate to total saturation in under two years. In response, twe…