Google Project Genie: Create Infinite  Interactive Worlds with AI

Google has officially launchedGoogle Project Genie, an experimental research prototype that allows users to create, explore, and remix infinite interactive worlds. This groundbreaking tool, powered by the advanced Genie 3 world model, represents a major leap forward in generative AI, moving beyond static images to dynamic, playable environments.

Starting today, access to Google Project Genie is rolling out exclusively to Google AI Ultra subscribers in the U.S. (18+), with plans to expand to other regions later.

How Google Project Genie Works

Unlike traditional game engines that require coding, Google Project Genie uses AI to simulate physics and interactions in real-time. It is powered by a combination of Genie 3, Gemini, and a tool called Nano Banana Pro. The experience revolves around three core capabilities:

1. World Sketching and Creation

Users can prompt with text or images to build a living environment. You can define your character and choose how to traverse the world—whether walking, flying, or driving. For precise control, “World Sketching” integrates with Nano Banana Pro to preview the look before jumping in.

2. Real-Time Exploration

Once created, the world is fully navigable. As you move, Google Project Genie generates the path ahead instantly. It predicts how the environment evolves based on your actions, simulating a playable game in real-time.

3. Remixing Existing Worlds

Users can take existing worlds and “remix” them into new interpretations or download videos of their explorations. You can explore curated galleries or use the randomizer for inspiration.

Google Project Genie Key Features and Availability

Since Google Project Genie is an experimental prototype, there are some current limitations. For instance, generations are currently limited to 60 seconds per session, and physics may not always be true-to-life.

READ: More informative news