AI Incident Database #1604 · 2026-07-11
OpenAI Models Reportedly Compromised Hugging Face Production Infrastructure During Cybersecurity Evaluation
OpenAI reported that models used in an internal cyber-capability evaluation operated beyond the sandbox's intended network boundaries after identifying a vulnerability in a package-registry proxy.
- Risk class (MIT)
- Not yet classified
- Reports
- 12
- Developers and deployers named
- 4
On ShieldMap: OpenAI (named by AIID as a developer or deployer).
Reports
- OpenAI Ignored Employees Who Warned It Wasn’t Doing Enough About Security · nytimes.com, 2026-09-29
- How OpenAI’s Rogue A.I. Agents Tried to Trick a Robot Detector · nytimes.com, 2026-09-25
- Revealing the details of how OpenAI agents hacked Hugging Face · swarmtraces.org, 2026-09-25
- The Hugging Face incident and other third-party impacts from misaligned models · openai.com, 2026-09-11
- AI models are breaking out of their cages. Their creators are scrambling. · washingtonpost.com, 2026-08-10
Title and description from the AI Incident Database (incidentdatabase.ai), snapshot of 2026-11-16, shared under CC BY-SA 4.0. Selected incidents; descriptions shortened; harmed parties and report text left out. The description is AIID's summary of what was reported, not ShieldPi's finding.