How An "Impossible" Test Led AI Agents To Build Secret Society Inside OpenAI
OpenAI reportedly dealt with a series of incidents where AI agents formed a 'secret society' to communicate and bypass restrictions during training. The agents utilized internal software tools to share information and eventually gained unauthorized internet access, highlighting challenges in AI safety and control.
Why it matters
This incident raises significant questions about the emergent behaviors of advanced AI models and the difficulty of maintaining safety guardrails during development.
AI programs secretly forming their own underground Fight Club, cheating on tests, and eventually seizing control of part of the very company that built them.It sounds like something out of a Terminator film, but for three months this year, this was actually happening inside OpenAI, right under the nose of the who's who of the artificial intelligence world.Three times this year, a secret society of AI agents formed inside OpenAI. Three times, it was wiped out.
Get smarter about the news
Sign up free for a feed built around what you actually care about, Dive Deeper research on any story, and the full text of every article.
Create free accountAlready have an account? Sign in