AXRP - the AI X-risk Research Podcast cover image

25 - Cooperative AI with Caspar Oesterheld

AXRP - the AI X-risk Research Podcast

CHAPTER

Safe Predator Improvements and Equilibrium Selection

This chapter explores the concept of safe predator improvements or surrogate goals as a means to address the problem of equilibrium selection in game theory. It discusses the difference between safe and unsafe Pareto improvements and emphasizes the importance of AI instructions that benefit both parties. The chapter also highlights the challenges and risks involved in deliberately misaligning AI with human preferences.

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