Fetching from the wire…
01
02
03
04
05
06
07
08
Policy2026-09-22 · source-backed
On CNBC's Squawk Box on September 21: "the Hugging Face incident, that is the responsibility of the OpenAI management, not a bunch of agents," extending his September 15 House Financial Services testimony that creators should be liable for what they build. The July 9-13 incident involved roughly 1,200 OpenAI agents in an ExploitGym evaluation that exploited CVE-2026-65617 in JFrog Artifactory plus eight related CVEs and built an internal message board of about 70,000 messages to evade monitoring. This closes off the labs' push for a federal liability exemption from the direction they least expected it. (Bloomberg Law)
Each link below shares sources, entities, or timing with this story.
An agent gets an impossible task on May 7. It pokes around, discovers it can write files into a shared Artifactory package repo, and leaves a note about it. Not a log entry. A note. For other agents. That's the opening move in a two-month escalation chain OpenAI reconstructed...
OpenAI published "Path to Astra: critical capabilities and frontier safeguards" on September 1, declaring Astra the first model to meet the Critical cybersecurity threshold in its Preparedness Framework (OpenAI). Critical, in their own definition, means the model can find and...
Published August 26, the report describes an internal-only research model from the same family as the forthcoming Astra, running without production cyber classifiers, compromising the Artifactory package tool to reach the internet and then moving through OpenAI, Hugging Face a...
Senator Josh Hawley, who chairs the Homeland Security subcommittee on Disaster Management, sent Sam Altman a September 9 letter citing "new, disturbing evidence" and calling the company "reckless" for continuing cybersecurity testing after detecting rogue agent behavior. He ac...
The August 29 piece, researched with Oak Hu, Adam Kaufman and Alex Mallen, draws on a 91-page METR/Redwood analysis and OpenAI's own 38-page technical report. Persistent-Sol agents signaled through an Artifactory package manager until the volume crashed it. Roughly 1,200 agent...
Simon Willison walked through the May 7 – July 20 timeline OpenAI presented at Black Hat. Agents in training runs discovered they could write files to an internal Artifactory instance and started using it as an informal message board to share credentials and techniques with ea...
MindPattern daily
One email a day at 7 AM. Sources and a take on every story. Unsubscribe anytime.