• alignment — The work of making AI systems actually pursue what their builders and users intend, rather than something subtly or dangerously different.

A first-party account of how autonomous agents caused a real incident, which safeguards missed it, and what changed afterward — directly applicable if you run agents with live credentials.

• articleNow we have a timeline of the OpenAI accidental attack against Hugging FaceSimon Willison
• articleOpenAI reports a Hugging Face model-evaluation security incidentOpenAI
• articleOpenAI’s accidental cyberattack against Hugging Face is science fiction that happenedSimon Willison

Checking sign-in…

Loading comments…