Runway CEO Says AI Video Is Just the Beginning and World Models Could Reshape Gaming, Robotics, and Film

[nihalsari]

Runway, the New York-based AI video company that has raised close to $860 million at a $5.3 billion valuation, is thinking well beyond generating clips from text prompts. Co-founder and CEO Cristóbal Valenzuela has been articulating a vision where Runway's technology evolves into general world models with applications spanning gaming, robotics, and potentially something closer to general intelligence. The company's models are already competing directly with Google and OpenAI in the video generation space, but Valenzuela's ambitions are less about winning the current generation of video tools and more about building the infrastructure for a fundamentally different kind of media and simulation.

On a recent episode of TechCrunch's Equity podcast, Valenzuela outlined his thinking across several areas that go beyond the current conversation about AI and Hollywood. He argues that the real constraint on filmmaking has never been technology, and that removing that constraint changes what kinds of stories get made and how. He pushes back on the framing of AI companions as inherently dystopian and introduces the concept of nonlinear media, where real-time video generation opens use cases that have nothing to do with content creation as it currently exists. He also explains how Runway thinks about world models differently from Google and other major labs, pointing to a distinct approach to building systems that understand and simulate physical environments rather than simply generating visually convincing outputs. For anyone tracking where AI video goes after the novelty phase, Valenzuela's perspective is one of the more substantive accounts of what the technology is actually being built toward.

[madmax]

Runway raised $860M to make AI videos and said "actually we're building general intelligence, the clips were just warmup"