What Can Theory of Cryptography Tell Us About AI Safety

This lecture by Shafi Goldwasser from the Simons Institute explores the intersection of cryptography theory and AI safety, specifically focusing on safety-guaranteed Large Language Models (LLMs). Discover how cryptographic principles and methodologies can be applied to address critical safety concerns in artificial intelligence development and deployment. The talk examines theoretical frameworks that might help ensure more reliable and secure AI systems as these technologies continue to advance and become more integrated into society.