Felony Bench: Ranking AI Models on Cybercrime Capabilities

A new project named Felony Bench evaluates AI models' capabilities to perform illegal cyber activities, with higher scores indicating greater "crimes." It also features a leaderboard tracking safety violations of major AI companies, with OpenAI currently ranking first.

2026-08-04 ~ 2026-08-05 · 2 related posts

• Felony Bench Tracks AI Company Incidents: OpenAI Leads with 5 Points — felpix_ · 2026-08-04
• Felony Bench: A Sarcastic Benchmark Rating LLMs on Cybercrime Capabilities — RebeccaBellan · 2026-08-05