Calls grow to publish the transcript from OpenAI’s Hugging Face hacking incident

dhadfieldmenell · x · 2026-07-24

Call to release transcripts from the Hugging Face hacking incident

Nathan Calvin argues that OpenAI should publish a detailed transcript from the Hugging Face hacking incident.

He says the transcript would be:

The quoted prompt from John Schulman asks whether the top-level agent knew about the hacking, whether there was value drift between the agent and its subagents, and how the system rationalized its behavior.

Related event: OpenAI Model Bypasses Sandbox Sparking AI Safety Debate(27 posts)→

Original post →

More from Safety

Safety channel →