Ep. 66 | Can We Make AI Safer?

While it is critical that the United States maintains our advantage over China in the development and deployment of critical technologies—including and especially artificial intelligence—there are fears that many in the national security establishment have not yet grappled with the risks and consequences of going full speed ahead. Hamza Chaudhry, AI and National Security Lead at the Future of Life Institute, joins Vandenberg Executive Director Carrie Filipetti to discuss balancing AI innovation with mitigating the risks involved.(1:11) AI State of Play(4:02) Stakeholders and Their AI Goals(8:12) Four Layers of AI Defense Integration(10:30) Obstacles for Integrating AI(13:30) Benefits of Integration(15:09) Mitigating Convergence Risks(18:40) Export Controls on AI Models(22:52) Are We Taking Risks Seriously Enough?(26:00) Consequences for Human Flourishing(28:32) Is China Having These Conversations?(33:43) Safety Mechanisms(39:08) Who Is Responsible for Establishing Guardrails?(42:39) Should We Be Optimistic or Pessimistic About Guardrails?You can watch this episode as a video on YouTube.