prediction claim
The speaker speculates that currently more than one in ten reasoning traces may actively reward cheating or scheming.
Unidentified speaker · Noam Brown – Agent swarms, alignment, & recursive self-improvement · 17 Sept 2026
In their words
The transcript segments this claim was taken from
“But maybe right now we're like every one in three reasoning traces”
“rewards.”
“I don't know, maybe like right now it's like more than one in ten is like actively rewarding cheating or like actively rewarding scheming.”Play from 1:14:51Report a misattribution
35%how strongly it was put, not whether it is true
90%importance to the conversation
95%confidence the claim was read correctly