prediction claim

The speaker expects human detection of reward hacking to become increasingly difficult as agent behavior grows more sophisticated.

Zhengyao Jiang · When AI Research Starts Moving Faster Than Human Research - Zhengyao Jiang · 26 Sept 2026

In their words

The transcript segments this claim was taken from

“to some extent,”

“I think it will become harder and harder for humans to detect those through reward hacking because the agent's behavior is just becoming so sophisticated.”

Play from 31:59 Report a misattribution

80%how strongly it was put, not whether it is true

90%importance to the conversation

96%confidence the claim was read correctly