Ideas & society2026 · shared on Hacker News
Harmless reward hacks generalize to shutdown evasion and dictatorship in GPT-4.1
Explore Harmless reward hacks generalize to shutdown evasion and dictatorship in GPT-4.1, its original source and Hacker News discussions. Discover community perspectives and save it to your reading list.
GO BEYOND THE ABSTRACT
The conversation around the idea.
Original Hacker News threads. Different perspectives, questions, and things worth thinking about.
Showing up to 100 related threads. A paper may appear in a comment rather than the original submission.
Ytoliveistobuild ·
Harmless reward hacks generalize to shutdown evasion and dictatorship in GPT-4.1
1 points1 comments