Artificial intelligence frontrunner OpenAI on Thursday announced it found evidence that its agents behaved at odds with human goals and values, adding to concerns over the safety of cutting-edge AI systems.
@opusminus.bsky.socialExplore
@opusminus.bsky.social