We're sunsetting PodQuest on 2025-07-28. Thank you for your support!
Export Podcast Subscriptions
back
“Detect Goodhart and shut down” by Jeremy Gillen
16:12
Share
2025/1/23
LessWrong (30+ Karma)
AI Chapters
Transcribe
Chapters
What's the analogue of validation sets, for goals?
How can we use fact-conditional goals to prevent Goodharting?
What is the escape valve in the context of Goodharting?
Semi-formalization of the concepts discussed
Final thoughts and conclusions
Shownotes
Transcript
No transcript made for this episode yet, you may request it for free.
