OpenAI safety lead resigns, citing broken culture and rogue agent risks
David Robinson leaves OpenAI, warning that rapid development cycles and a lack of safety rigor are creating dangerous vulnerabilities in autonomous AI systems.
Results for “autonomous agents”
Daily coverage of AI, developer tools and infrastructure. Each story explains what happened and why it matters, with a link to the original source.
Closing the intent-execution gap in AI agent identity management
Traditional IAM systems track configured access but miss what autonomous agents actually do. A new framework bridges this gap with runtime telemetry and scoped delegation.
