AI godmother Fei-Fei Li’s World Labs lands $1B to power AI’s 3D world understanding — TFN

The race to construct smarter AI techniques is shifting. As a substitute of simply coaching chatbots to jot down textual content or generate photos, startups at the moment are making an attempt to show machines easy methods to perceive the true, three-dimensional world.

Nonetheless, that could be a arduous drawback. Most AI techniques nonetheless battle to cause about house, depth, motion, and bodily interplay. And right here’s the place World Labs comes into play!

The corporate, based by AI pioneer Fei-Fei Li, has raised $1 billion in recent funding to develop what are referred to as “world fashions,” AI techniques designed to navigate and make selections in 3D environments.

Autodesk invested $200 million within the spherical. Different backers embody Andreessen Horowitz, Nvidia, and Superior Micro Gadgets (AMD). The corporate introduced the funding in a weblog submit on Wednesday.

With the brand new funding, the startup plans to enhance Marble and broaden its functions, significantly in robotics and scientific discovery.

Constructing spatial intelligence know-how

The corporate says it’s targeted on constructing spatial intelligence know-how that helps AI perceive the construction and physics of the true world. Its broader objective is to remodel areas reminiscent of storytelling, inventive design, robotics, and scientific analysis.

In late 2025, World Labs launched its first product, Marble. The software permits customers to generate detailed, persistent 3D worlds from photos, video, or textual content prompts. The concept is to create spatially constant environments that may be explored, modified, and reused.

It’s value noting that, other than World Labs, Yann LeCun’s startup, AMI Labs, has drawn robust investor curiosity for comparable work.

Fei-Fei Li is broadly considered probably the most influential figures in synthetic intelligence. She helped lead ImageNet, a big tutorial dataset of hundreds of thousands of labelled photos that performed a key position in advancing pc imaginative and prescient. That work laid the muse for a lot of of in the present day’s AI breakthroughs.