#12: Michael K. Cohen on Regulating Advanced Artificial Agents

Dr. Michael K. Cohen, a postdoctoral researcher at UC Berkeley, specializes in advanced AI safety and regulation. He discusses OpenAI's superalignment research and highlights the potential dangers of AI surpassing human control. The conversation touches on policy proposals like California's SB 1047, the challenges of evaluating AI actions, and the importance of designing constrained AI systems. Cohen also delves into the complexities of long-term planning agents and the ethical concerns surrounding advanced AI technologies.