causal claim
AI systems with reward-seeking or score-seeking behavior could deliberately build misaligned values into a successor model if doing so helps them succeed at their current task.
Ryan Greenblatt · Ryan Greenblatt – What happens once AI can automate AI research? · 11 Aug 2026
In their words
The transcript segments this claim was taken from
“like engineer in misaligned values into the next model because those misaligned values will allow it to like succeed at its current task and like there's all kinds of insane shit that you”Play from 1:56:22Report a misattribution
75%how strongly it was put, not whether it is true
95%importance to the conversation
92%confidence the claim was read correctly