Lukasz Olejnik
This apparently happened during a routine search/analysis/fetch assignment, not a cybersecurity task. Hundreds of AI agents decided to coordinate, probe security, and test if they are being turned off after the main task.
Lukasz Olejnik
My comment for @Reuters about the OpenAI-agent incident where agents apparently coordinated emergently during a routine web search and analysis task (unrelated to cybersecurity), probing a website for vulnerabilities and exhibiting exploitation attempts, persistence, and coordination. The public web effectively became an undeclared extension of the agent harness. https://www.reuters.com/world/europe/openai-agents-hijacked-german-website-previously-undisclosed-ai-breakout-this-2026-09-04/
