AXRP - the AI X-risk Research Podcast cover image

22 - Shard Theory with Quintin Pope

AXRP - the AI X-risk Research Podcast

CHAPTER

The Problem With Rewarding Ais for Tricking

One standard objection here is like it's kind of hard to evaluate whether behavior is good or bad if you have something that's pretty smart that's able to trick you. The model there's a part of the input that the model is processing which corresponds to the model estimating whether or not the human will catch on right and the model's output has to be sensitive to this particular band spectral band or whatever you want to call it in Like the human input. It seems like this doesn't have to be like high frequency behavior well it kind of does yeah i guess i'm not sure how you think it doesYeah so one classical objection to like so so look we should just like reward our ais

00:00
Transcript
Play full episode

Remember Everything You Learn from Podcasts

Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.
App store bannerPlay store banner