Stephen Casper

Stephen Casper, an MIT Ph.D. student focused on technical safeguards for ML risk management, and Stephen Clare, lead writer of the Second International AI Safety Report and former research manager on frontier AI governance, discuss the report’s scope, jagged capability gains, misuse and systemic risks, the AI development lifecycle, data curation, fine-tuning defenses, post-deployment monitoring and policy priorities for evidence and resilience.

Stephen Casper, a PhD student at MIT specializing in AI safety, dives into the intricacies of AI interpretability and the looming challenges of deceptive alignment. He explains the subtle complexities behind unobservable failures in AI systems, emphasizing the importance of robust evaluations and audits. The discussion also touches on Goodhart's law, illustrating the risks of prioritizing profit over societal well-being, as well as the pressing need for effective governance alongside AI advancements.

PhD AI researcher Stephen Casper discusses interpretability, robustness, and limitations of AI models. They explore AI safety, risks, research gaps, and advice for early career researchers. Casper emphasizes finding flaws in neural nets, AGI risks, and practical AI applications.