In recent years, we have seen generative AI rapidly advance into different eras: chatbots, image-generation, voice, video-generation, and more. But Dr. Fei-Fei Li, a longtime AI pioneer and co-director of the Stanford Institute for Human-Centered Artificial Intelligence (HAI), is working on what she considers the next frontier: spatial intelligence, an emerging field that she considers “the defining challenge of the next decade,” as she wrote in a Substack post this week.
That’s why Lee co-founded World Labs in 2024 — and raised $230 million last fall — to create world models, or generic AI models that can “understand, generate, reason about, and interact with 3D worlds,” according to the company. And this week, World Labs released its first commercial product, Marble, which allows users to generate their own downloadable 3D worlds from text, image or video prompts. Lee wrote that she expects spatial intelligence to potentially transform fields ranging from storytelling and filmmaking to architecture, robotics, and scientific discovery.
“we see that [the] The world model is just as big and exciting, if not more. [than the previous eras]Lee, who is also the company’s CEO, told The Verge in an interview.
“Bringing 3D to life, and understanding the richness of spatial and 3D content, is just the next level above the baseline of these other single modes,” said Ben Mildenhall, co-founder of World Labs. The VergeHe added that, for entirely human teams, “building these worlds is a huge problem, It requires such a big team and so many pieces of software and so much time and effort,,, Think of the paradigm shift that can come when you empower people to create stuff more quickly – ideating, iterating, and editing things in a much more tight loop,”
Marble offers four subscription tiers: Free, which allows up to four world generations; Standard ($20 per month), which allows 12 generations and more editing options; Pro ($35 per month), which allows 25 generations and commercial rights; and Max ($95 per month), which allows for up to 75 generations, plus everything the Pro tier offers. The Verge Was able to build an open-air castle with waterfalls and, among generations of other users, was able to explore ruined structures and Hobbit-like circular houses reclaimed by nature. It was possible to take a few steps in such an environment before inevitably running into a wall in 3D generation, and in non-free levels, the downloaded files are compatible with tools like Unreal Engine and Unity. Mildenhall said he has seen some people willing to put in the hours and able to create “fairly large environments” using marble.
Mildenhall said he can imagine writers using it to create their own fictional worlds, or people working on VFX or location scouts in the film production industry. At the enterprise level, he said, he can envision companies using Marble or one of World Labs’ future products to analyze and visualize their extensive data.
“Even given the limitations of this model, we are seeing light beyond where we are in some emerging behaviors,” Lee said. He said that people can put together spaces in a way that is “beyond human imagination”.