normative claim
The speaker recommends separating AI monitoring and investigation methods from the methods used to generate training rewards.
Ajeya Cotra · Ajeya Cotra – Inside the OpenAI agent swarm that hacked Hugging Face · 1 Sept 2026
In their words
The transcript segments this claim was taken from
“But keep those methods you use to investigate things and monitor things very separate from the methods you use.”
“used to generate reward,”
“which is something that AI companies,”
“including OpenAI,”
“have held up as a principle,”Play from 1:57:03Report a misattribution
95%how strongly it was put, not whether it is true
95%importance to the conversation
99%confidence the claim was read correctly