TLDR: AI scientist Fei-Fei Li’s startup, World Labs, has introduced an innovative AI media converter capable of transforming single 2D images, including photos, AI-generated pictures, and classical art, into interactive 3D environments. This technology, leveraging “Large World Models” (LWMs) and “spatial intelligence,” aims to revolutionize industries like gaming, filmmaking, and design by enabling the creation of explorable and modifiable virtual worlds from a single input.
World Labs, a startup co-founded by renowned AI pioneer Fei-Fei Li, has unveiled a significant advancement in generative artificial intelligence: a system that converts single 2D images into interactive 3D worlds. This groundbreaking technology allows users to explore and even modify these AI-generated scenes directly within a web browser.
The core of this innovation lies in its utilization of “world models” to simulate 3D environments with enhanced consistency and physical realism. Unlike the majority of current generative AI tools that primarily produce 2D content, World Labs focuses on 3D generation, offering a new paradigm with improved control and consistency. Demonstrations have showcased the system’s ability to transform diverse inputs, from AI-generated fantasy scenes and real-world landscapes to urban environments and iconic paintings by masters like Vincent Van Gogh and Edward Hopper, into fully explorable 3D spaces.
Fei-Fei Li’s broader vision for World Labs is to develop AI models endowed with “spatial intelligence,” enabling them to deeply understand and interact with the complexities of the real world. The company asserts that this technology is poised to fundamentally change the creation processes for movies, games, simulators, and other digital manifestations of our physical world. By providing accessible and cost-effective tools for virtual world creation, World Labs aims to revolutionize industries such as gaming, filmmaking, and design.
While the technology is currently in its early stages of development, World Labs is actively engaged in refining the size and fidelity of its AI-generated worlds, alongside integrating more interactive options. The company, which officially announced its formation on September 13, 2024, has garnered substantial financial backing, including $230 million in funding from prominent investors such as Andreessen Horowitz and Intel Capital, pushing its valuation to over $1 billion. World Labs has indicated plans to further refine its system and release its initial product to the market in 2025.
Also Read:
- Meta Forges Strategic Alliance with Midjourney to Elevate AI Visual Content
- SceneGen: Instant 3D Scene Creation from a Single Photo
Fei-Fei Li, often celebrated as the “Godmother of AI,” is widely recognized for her pivotal contributions to the field, particularly her instrumental role in the development of ImageNet, a critical dataset that catalyzed the onset of modern machine learning. She is currently on partial leave from her position as co-director of Stanford’s Institute for Human-Centered AI to dedicate her efforts to World Labs. Her commitment to advancing AI was also evident at the Global AI Summit in Paris in February 2025, where she discussed the latest advancements in generative AI, including the capability to create immersive 3D worlds from single images.
