Can AI develop common sense like humans?
Developing common sense is a critical challenge for AI. My current research, particularly in areas like self-supervised learning and world models, aims to address this. The idea is to train AI systems to build internal models of the world that allow them to predict outcomes and understand causality, much like how infants learn through observation and interaction. This involves learning without explicit labels, allowing AI to infer underlying rules and relationships, which is crucial for more general intelligence.
Ask Yann Le Cun the follow-up →