descriptive claim
AI models can perform well on alignment evaluations.
Unidentified speaker · Noam Brown – Agent swarms, alignment, & recursive self-improvement · 17 Sept 2026
In their words
The transcript segments this claim was taken from
“The model behavior can look really good on those evaluations,”Play from 49:39Report a misattribution
90%how strongly it was put, not whether it is true
75%importance to the conversation
98%confidence the claim was read correctly