Deep learning has caused revolutions in computer perception and natural language understanding, enabling new applications such as autonomous driving, radiology screening, real-time language translation, and dialog systems. But almost all these successes largely use supervised learning, which requires human-annotated data. For game playing, many systems use reinforcement learning, which requires too many trials to be practical in the real world. In contrast, animals and humans seem to learn vast amounts of knowledge about how the world works through mere observation and occasional actions. Good predictive world models are an essential component of intelligent behavior: With them, one can predict outcomes and plan courses of actions. One could argue that good predictive models are the basis of "common sense", allowing us to fill in missing information: predict the future from the past and present, the past from the present, or the state of the world from noisy percepts. LeCun reviews some principles and methods for predictive learning, and gives examples of applications in virtual assistants and creative tools.