The story’s score and claims

Comments on No. 46 of 122 · GitHub · edition of Tuesday 17 November 2026

SatBench a Benchmark of AI Benchmarks

Summary

SatBench, a project by Ethan Mollick, treats 173 frequently cited AI benchmarks as questions on one exam, counting one solved once the best reported score reaches 85%. It stands at 63% solved, up about 20 points in the past year.

Comments

Ranked by a public rubric, not by votes: what a comment adds counts, not who wrote it. How comments are ranked · Comment rules