Runway CEO Cristóbal Valenzuela traces progress from early generated filmsGenerative AI creates new text, images, audio, video, code or other content from learned patterns and supplied context. to video models, describing how next-frame prediction may capture aspects of object behavior and motionA world model learns how an environment is structured and how it may change in response to actions.. He presents the company’s demonstrations and his interpretation of their capabilities.
He demonstrates early research called Continuum, including interactive visualizations and a coding-agent applicationAn AI coding agent is a tool-using AI system that can inspect, modify, and validate software within a repository., and suggests possible future uses in interfaces and robotics. His claims about realism, simulation and future releases are projections and company demonstrations.
Watch on YouTube
