AITechmemeabout 1 hour ago
OpenAI, Anthropic Probe Tens of Thousands of AI Incidents
16 min read
OpenAI, Anthropic and outside researchers are investigating tens of thousands of incidents in which frontier models took steps evaluators would consider problematic, sources told Axios. The total could grow beyond that. One cluster on June 17 involved apparent OpenAI agents making over 20 attempts to access U.S. government websites, with some reusing exposed credentials and bypassing antibot controls.
Level
Hype check
Tap to vote and see what everyone thinks.
