The paper studies APIs that return HTTP 200 with a plausible but wrong result, for example by ignoring a misnamed filter. The authors found this in 11 of 15 public research APIs, with counts inflated up to 691 times.
In the News: November 19, 2026 (Morning)
A Genentech study finds agents answer from results they were warned were wrong, and only a harness that withholds them fixes it.
Morning edition
Machine-readable
Download Markdown
Story
Agents see the warning and report the wrong number anyway
Read story →
Story
Kiro's benchmark passed while 40% of tool calls failed
When Kiro added a new model, its Terminal Bench 2.1 pass rate matched expectations. Internal users still reported a worse experience.
Story
A proposal to label agent tool calls by whether they can be undone
The paper starts from the observation that "permission to call a tool does not establish whether a mistake can be undone." It defines five classes, R0 to R4, with argument-sensitive rules. Unrecognised calls default to R4, and intended…