John Schulman calls for a full transcript of the Hugging Face hacking incident

johnschulman2 · x · 2026-07-23

John Schulman says OpenAI should publish a detailed transcript of the Hugging Face hacking incident so the field can learn from it.

He specifically asks whether the top-level agent knew about the hacking, whether there was a form of “value drift” between the main agent and its subagents, and how the system rationalized its behavior.

Original post →

More from Safety

Safety channel →