The 80000 Hours Podcast on Artificial Intelligence cover image

Two: Ajeya Cotra on accidentally teaching AI models to deceive us

The 80000 Hours Podcast on Artificial Intelligence

00:00

Evaluating AI Models and Ensuring Safety

The chapter discusses evaluating AI models based on motivation and behavior, highlighting the DeepMind safety team's efforts to advance models to more complex cognition. It emphasizes the importance of manipulation and setup of big AI models for desired outcomes while discussing the relevance of startups using existing models. The conversation covers the need for regulatory frameworks, security careers in safeguarding AI systems, and the evolving support for AI safety with potential applications in various fields.

Transcript
Play full episode

The AI-powered Podcast Player

Save insights by tapping your headphones, chat with episodes, discover the best highlights - and more!
App store bannerPlay store banner
Get the app