← Home
No Priors · September 25, 2025 · 50m
Dan Hendrycks: The Real Risks of Advanced AI
CAIS executive director Dan Hendrycks discusses the real risks of advanced AI: not science fiction scenarios but concrete near-term dangers from misuse, misalignment, and concentration of power.
This summary was generated from show notes and public descriptions, not from a full transcript review. Details may contain inaccuracies.
Canon
•
Hendrycks frames AI safety as environment design: building evaluation frameworks, safety standards, and monitoring systems into the AI development environment shapes safer development behavior.
•
The Stoic safety researcher — controlling evaluation rigor, accepting uncertainty about future risks
Hendrycks applies Stoic reasoning to AI safety work: he can control the rigor of his safety evaluations and benchmarks but cannot control the pace of AI capability development or how systems will be used.Was this useful?