Skip to content 
Key Takeaways
- Overview: World Labs, founded by Fei-Fei Li, unveiled AI technology to create interactive 3D scenes from a single image.
- Features: Users can explore scenes with depth-of-field effects, modify object colors, and adjust lighting, all with stable physics.
- Applications: Aimed at revolutionizing gaming, filmmaking, and design by simplifying 3D world creation.
- Funding: Raised $230M, valued over $1B, with plans for product launch by 2025.
World Labs, a startup founded by AI visionary Fei-Fei Li, has introduced its debut project: an innovative AI system capable of generating interactive 3D scenes from a single image. Unlike existing technologies that create 3D models from photos, World Labs’ solution brings these scenes to life with interactivity and customizability.
Step into the Scene
“Our technology allows you to step into any image and explore it in 3D,” World Labs shared in a blog post. The system generates everything beyond the input image, creating immersive environments that users can navigate using a keyboard and mouse.
Available as a demo on World Labs’ website, the AI-generated scenes offer features such as:
- Interactive exploration: A controllable camera with adjustable depth of field (DoF) for realistic focus effects.
- Dynamic interactivity: Users can modify object colors and adjust lighting within the scenes.
The scenes maintain a sense of solidity and depth while adhering to basic physics principles, setting them apart from existing world models that struggle with consistency or resolution issues.
Advancing World Models
World Labs’ technology belongs to a cutting-edge AI category known as “world models,” which simulate 3D environments. Unlike competitors, such as Decart’s Minecraft-simulating Oasis, which suffers from low resolution and memory issues, World Labs’ scenes remain stable once generated and ensure a higher level of realism.
The company believes this shift from 2D to 3D generative AI will revolutionize industries. “Generating in 3D improves control and consistency,” World Labs wrote. “This will transform how we create movies, games, simulators, and other digital representations of our physical world.”
Room for Growth
As impressive as the technology is, it’s still in its early stages. Current limitations include restricted movement areas and occasional rendering glitches, such as objects blending unnaturally.
“We’re hard at work improving the size and fidelity of our generated worlds,” the company assured.
Backed by Big Names
Founded earlier this year, World Labs has already secured $230 million in funding from top investors, including Andreessen Horowitz (a16z), Intel Capital, AMD Ventures, and Eric Schmidt. With a valuation exceeding $1 billion, the startup aims to release its first commercial product by 2025.
Future Applications
World Labs envisions its tools being used across industries, from video game development to filmmaking. “Creating virtual, interactive worlds today requires enormous time and costs,” said co-founder Justin Johnson on the a16z podcast. “World models will enable not just static images or videos but fully simulated, vibrant, and interactive 3D worlds.”
As the company refines its technology, World Labs is positioning itself to redefine how we create and interact with digital environments.
View more details on the Word Lab’s features.
More from Earn Hustles Insider:
Connect with us on LinkedIn to expand your network and grow together: Earn Hustles Insider.
