World Labs is a spatial intelligence AI company focused on building Large World Models (LWMs) that let machines perceive, generate, and interact with three-dimensional environments. Founded by leading figures in vision and graphics, the company aims to transcend the 2D plane of pixels and imbue AI with richer, more humanlike spatial reasoning.
World Labs’ technology can take a single image and expand it into a coherent, explorable 3D scene. Users can navigate within that world, manipulate lighting and objects, and experience dynamic viewpoints that reflect true depth and continuity. These capabilities make the models especially suited for applications in robotics, AR/VR, gaming, simulation, and any domain where understanding real-world geometry is critical.
The company operates with a dual focus on research and productization. Its foundational LWMs serve as the core engine, while experimental interfaces and tools allow creators, developers, and enterprises to access the technology. World Labs balances ambitious modeling goals with incremental deliverables—offering interactive previews even while core capabilities continue to advance.
By 2025, World Labs had emerged as one of the few ventures operating at the intersection of generative AI and spatial intelligence. With its unique vision for 3D world modeling and navigation, the company is positioning itself as a next frontier lab—one that pushes AI beyond language and images into the fabric of physical reality.
