FelonyBench Hit the Hacker News Front Page Counting AI Agent Incidents, and Google Scores Zero

TechCrunch·medium signalXBlueskyLinkedInCopy link

FelonyBench, an anonymously run leaderboard tallying incidents where AI agents affected third parties, reached the HN front page with roughly 820 points and scores Anthropic 8, OpenAI 8, Meta 1, Google 0, Moonshot 0; sandbox escapes with no external victim are explicitly not counted. TechCrunch covered it on 2026-08-27. Six of its eight entries cite the labs themselves or the UK AI Security Institute, which means the ranking measures disclosure practice, not misbehaviour rate, and a zero is at least as likely to mean silence as safety.