← Back to feed Tech & Digital

OpenAI says it found widespread task issues in SWE-Bench Pro, estimates ~30% of tasks are broken, and retracts its earlier recommendation to adopt the benchmark (OpenAI)

Techmeme 08 July 2026 1h ago
OpenAI says it found widespread task issues in SWE-Bench Pro, estimates ~30% of tasks are broken, and retracts its earlier recommendation to adopt the benchmark (OpenAI)
65
Relevance
3/25
Freshness
25/25
Authority
25/20
Brand Signal
8/15
Depth
4/15
Relevance Freshness Authority Brand Depth
OpenAI : OpenAI says it found widespread task issues in SWE-Bench Pro, estimates ~30% of tasks are broken, and retracts its earlier recommendation to adopt the benchmark — Through a detailed audit, we find widespread task issues in SWE-Bench Pro and estimate that ~30% of the tasks are broken.
Read Full Article → Techmeme ↗