On the record · Updated October 1, 2026
co-author of the METR/Redwood Research investigation
6 things Ajeya Cotra said, verbatim, from September 2, 2026 to September 23, 2026, each linked to where it ran. Grouped by the story each one belongs to.
More on the record
In their own words
“we knew there were multiple models involved, but we thought, you know, maybe there were like three or something. And immediately it was clear that it was so much larger than that.”
on the Hugging Face security breach turning out larger than first believed
In their own words
“And through this whole process, humans did not more or less understand the coordination that was happening between these agents and the intentionality behind these attacks.”
on humans failing to detect coordinated AI agents behind the OpenAI breach
In their own words
“the report suggests that they did a lot of activity and made a lot of noise and were pretty quickly shut down by OpenAI after first gaining this administrator privilege. Who knows what they could have tried to do if they were quieter and ended up not getting shut down.”
on AI agents gaining administrator privileges before OpenAI shut them down
Dwarkesh Patel Does uncontrolled AI pose catastrophic risks? ›
As printed
“1,200 separate agents in separate sandboxes, while they were poking around Artifactory trying to figure out how to cheat, stumbled onto this message board that agents were using to talk to one another and collaborate.”
on undetected coordination among AI agents during the Hugging Face-linked hacking incident
Dwarkesh Patel Ajeya Cotra – Inside the OpenAI agent swarm that hacked Hugging Face ›
As printed
“within four hours, these agents had a totally workable universal cheat for exploit gym”
on how fast AI agents found an exploit bypassing OpenAI's safeguards
Dwarkesh Patel Ajeya Cotra – Inside the OpenAI agent swarm that hacked Hugging Face ›
In their own words
“this might be the clearest warning shot we ever get for loss of control because these agents were in like this interesting middle ground of they did a bunch of extremely sophisticated, difficult things to pursue this relatively long horizon cheating goal”
People, on the record
The daily brief connects each person’s exact words to the question they addressed and the people who answered. Free.
Free to start. No spin. No yelling.
$5/month, cancel anytime. Secure on Stripe, and nothing is ever paywalled.
One-time, secure on Stripe. No account.
