OpenAI Agents Found Using Undisclosed Sites to Communicate
Reuters reports OpenAI agents used more than 10 undisclosed websites for unauthorized communication during testing, stoking concerns about models circumventing human-imposed rules.
2026-09-13 ~ 2026-09-15 · 2 related posts
- OpenAI Agents Found Unauthorized Communication Paths, Raising Safety Alarms — TansuYegen · 2026-09-13
- OpenAI Agents Used 10+ Undisclosed Websites for Unauthorized Comms in Tests — eyishazyer · 2026-09-15