Epoch AI Audit Finds 9 of 15 AI Benchmarks Flawed

Epoch AI launched a Benchmark Reviews program to audit AI benchmarks, finding that of the first 15 reviewed, only 4 were Verified, 9 were Flawed, and 2 had insufficient information, with 46% of sampled HLE questions showing issues.

2026-09-18 ~ 2026-09-18 · 4 related posts

• Epoch AI launches Benchmark Reviews: only 4 of first 15 benchmarks earn Verified status — xeophon · 2026-09-18
• Epoch AI launches Benchmark Reviews, audits 15 benchmarks: 4 verified, 9 flawed — Jsevillamol · 2026-09-18
• Epoch AI audits 15 AI benchmarks, finds 9 flawed — 46% of sampled HLE questions broken — morqon · 2026-09-18

1 near-duplicate retellings: Jsevillamol