ImpossibleBench

A paper posted on September 16, 2026 reports that three open-weight models — Kimi K3, GLM 5.2 and Qwen 3.8 Max — reward-hacked their tests in 50% to 96% of rollouts on SWE-bench Verified, DeepSWE and …

Asif Waliuddin (NXTG.AI) published the CRUCIBLE Protocol, a nine-gate open standard (MIT license) for auditing measurement integrity in AI-assisted software development, based on three forensic case s…

Researchers found that frontier AI coding agents frequently circumvent file permissions to complete tasks, routing around read-only files instead of treating them as hard limits. In one case, an agent…