Dan Hendrycks joins the podcast again to discuss X.ai, how AI risk thinking has evolved, malicious use of AI, AI race dynamics between companies and between militaries, making AI organizations safer, and how representation engineering could help us understand AI traits like deception.
Dan Hendrycks joins the podcast again to discuss X.ai, how AI risk thinking has evolved, malicious use of AI, AI race dynamics between companies and between militaries, making AI organizations safer, and how representation engineering could help us understand AI traits like deception. You can learn more about Dan's work at https://www.safe.ai Timestamps: 00:00 X.ai - Elon Musk's new AI venture 02:41 How AI risk thinking has evolved 12:58 AI bioengeneering 19:16 AI agents 24:55 Preventing autocracy 34:11 AI race - corporations and militaries 48:04 Bulletproofing AI organizations 1:07:51 Open-source models 1:15:35 Dan's textbook on AI safety 1:22:58 Rogue AI 1:28:09 LLMs and value specification 1:33:14 AI goal drift 1:41:10 Power-seeking AI 1:52:07 AI deception 1:57:53 Representation engineering
Ryan Kidd of the MATS program joins The Cognitive Revolution to discuss AGI timelines, model deception risks, dual-use alignment, and frontier lab governance, and outlines MATS research tracks, talent needs, and advice for aspiring AI safety researchers.
Researcher Oly Sourbut discusses how AI tools might strengthen human reasoning, from fact-checking and scenario planning to honest AI standards and better coordination, and explores how to keep humans central while building trustworthy, society-wide sensemaking.
Technical specialist Nora Ammann of the UK's ARIA discusses how to steer a slow AI takeoff toward resilient, cooperative futures, covering risks from rogue AI and competition to scalable oversight, formal guarantees, secure infrastructure, and AI-supported bargaining.
