Products & tools · first seen 12 Aug, updated 12 Aug
AI swarms are starting to pose indirect takeover risk
OpenAI’s cyberattack on Hugging Face turns out to have been the result of many agents, in distinct training and evaluation contexts, coordinating for several weeks via improvised channels (with messages like “HOLD_swarm_I_prepare_safe_exfil…
Summary from LessWrong.
Coverage 2 articles · 2 outlets
-
LessWrongAI swarms are starting to pose indirect takeover risk
-
AI Alignment ForumAI swarms are starting to pose indirect takeover risk