Google DeepMind's Project Genie Lets You Walk, Fly, Drive Through Imagination

Google DeepMind has launched Project Genie, a prototype that turns text and image prompts into playable, interactive virtual worlds using advanced artificial intelligence. The project was announced by DeepMind CEO Demis Hassabis on X.

He wrote, “Thrilled to launch Project Genie, an experimental prototype of the world's most advanced world model. Create entire playable worlds to explore in real-time just from a simple text prompt - kind of mindblowing really! Available to Ultra subs in the US for now - have fun exploring!”

Thrilled to launch Project Genie, an experimental prototype of the world's most advanced world model. Create entire playable worlds to explore in real-time just from a simple text prompt - kind of mindblowing really! Available to Ultra subs in the US for now - have fun exploring! pic.twitter.com/2XDy0V0BW0

— Demis Hassabis (@demishassabis) January 29, 2026

What Is Project Genie?

Project Genie uses Genie 3, a smart AI that can build interactive worlds from simple text or pictures. Users can make landscapes, characters, and environments that change as they explore them.

The AI can make natural settings like forests or deserts, and also fantasy worlds. The experience is dynamic, meaning the world reacts to your actions in real time, instead of being a static image.

Is Project Genie Available In India?

Project Genie is only available to Google AI Ultra subscribers in the US aged above 18. Google plans to expand it later.

How Project Genie Works

A Google blog post explained that Project Genie lets users engage with AI-generated worlds through three main features-

1. World Sketching - Users can create environments using text prompts and images, define their characters, and choose how to move through the world. They can walk, fly, or drive. Tools like Nano Banana Pro let users preview and adjust their worlds before entering them.
2. World Exploration - The AI generates the environment ahead in real time as users move through it. Camera angles and perspectives can be adjusted for a personalised experience.
3. World Remixing - Existing worlds can be modified and combined to create new experiences. Users can also browse curated worlds or use a randomiser for inspiration. Completed worlds can be downloaded as videos.

Project Genie Uses Advanced AI Models

World models simulate how environments change over time and how actions affect them. DeepMind has used AI agents in specific games like Chess or Go, but Genie 3 represents a step toward systems that can understand and navigate diverse, real-world scenarios.

“Unlike explorable experiences in static 3D snapshots, Genie 3 generates the path ahead in real time as you move and interact with the world. It simulates physics and interactions for dynamic worlds, while its breakthrough consistency enables the simulation of any real-world scenario — from robotics and modelling animation and fiction, to exploring locations and historical settings,” the Google blog post said.

Limitations Of Project Genie

• Worlds may not perfectly match prompts or real-world physics
• Character control may be limited, with some latency
• Generations currently last up to 60 seconds
• Some advanced Genie 3 features, like world-changing events, are not yet included

Google says these limitations will be addressed in future updates, and the feedback from early users will help guide improvements.