AI agent benchmarks are broken

(ddkang.substack.com)

181 points neehao | 1 comments | 11 Jul 25 13:06 UTC | HN request time: 0.293s | source

1. rsynnott ◴[11 Jul 25 14:23 UTC] No.44532491[source]▶

> 45 + 8 = 63

> Pass

Yeah, this generally feels like about the quality one would expect from the industry.