Report
AI-risk interview by the lead author of a Google DeepMind control plan
A post calls the interview insightful and identifies its author as the lead author of dangerous-capability and scheming evaluation frameworks.
TLDR
A post shares an interview on AI risk and calls it insightful. It describes the interview as being by the lead author of a Google DeepMind control plan, a dangerous capability evaluation framework and a scheming evaluation framework.
Combined views
540
1 Source, first seen ago