Google Unveils Project Genie, an AI Tool That Lets Users Build and Explore Virtual Worlds

Google has introduced Project Genie, a new experimental AI-powered web application that allows users to design and explore their own interactive digital worlds. The tool blends creativity with artificial intelligence, enabling environments to be generated in real time from simple text prompts or images. Early demonstrations have already caught attention — including praise from Google CEO Sundar Pichai, who described the experience as “out of this world.”

Announcing the launch on X, Pichai revealed that the prototype runs on Google’s latest generation of AI models. He wrote, “Project Genie is a prototype web app powered by Genie 3, Nano Banana Pro + Gemini that lets you create your own interactive worlds.” Sharing a clip of an astronaut navigating a space station created through the tool, he added, “I’ve been playing around with it a bit and it’s out of this world:)”

At its core, Project Genie combines several Google AI systems. The Genie 3 model handles world generation, while Nano Banana Pro improves visual quality and rendering. Gemini, Google’s flagship AI platform, supports the overall intelligence behind the scenes, allowing environments to respond dynamically to user actions.

Using the platform is straightforward. Users begin with what Google calls “World Sketching.” By entering text descriptions or uploading images — even photos of everyday objects — the AI builds custom environments and characters. From there, users can explore their worlds from either a first-person or third-person perspective.

What sets Project Genie apart is its ability to create scenes on the fly. Instead of loading a prebuilt environment, the system generates landscapes and interactions in real time as users move through the space. The AI also simulates physics and object behaviour to make the experience feel more natural and immersive.

Beyond building from scratch, the tool lets users’ remix or modify existing worlds. This feature encourages experimentation and helps users quickly adapt environments with fresh prompts or ideas.

There are some limits, however. Each session currently allows only 60 seconds of exploration, and Google cautions that the worlds may not always appear fully realistic. Users might also encounter occasional latency when controlling characters.

For now, Project Genie is available exclusively to Google’s AI Ultra subscribers in the United States. According to Google DeepMind’s Diego Rivas, the limited rollout is intentional. He explained, “It’s really for us to actually learn about new use cases that we hadn’t thought about.”

Looking ahead, Google believes the technology could extend beyond entertainment or creativity. The company suggests that similar AI-generated simulations might one day help robots learn how to navigate real-world environments.

With Project Genie, Google signals its growing push into immersive AI experiences — and offers a glimpse into how digital creation could become faster, smarter, and more interactive than ever before.