Google Launches Project Genie: Sundar Pichai Calls It 'Out Of This World'

Google has introduced Project Genie, a new AI-powered web app that allows users to create and explore interactive virtual worlds. Designed as a creative sandbox, the tool lets users build environments that can be navigated in real time rather than just viewed as static visuals.

Google CEO Sundar Pichai revealed that he has been experimenting with the app himself, describing the experience as “out of this world.”

ALSO READ: Elon Musk Led xAI Opens Writing Specialist Role To Train Grok AI

Sundar Pichai Shares His Project Genie Experiment

Announcing the release on X, Sundar Pichai said that Project Genie is powered by Genie 3, Google’s AI world generation model that was previously released as a limited research preview.

“Project Genie is a prototype web app powered by Genie 3, Nano Banana Pro + Gemini that lets you create your own interactive worlds.”

Pichai also shared a clip created using the tool, showing an astronaut exploring a space station in Earth’s orbit. Reflecting on his experience with the app, he added, “I’ve been playing around with it a bit and it’s out of this world:)”

How Project Genie Lets You Build Your Own World

Project Genie begins with a step Google calls world sketching. Users can input text prompts or upload images, which the AI then uses to generate environments, characters, and complete virtual spaces. These images can even be based on real-world objects, such as toys, which are then transformed into explorable digital worlds.

Google has integrated Nano Banana Pro into the tool to enhance visual quality. Once a world is generated, users can explore it from multiple perspectives, including first-person and third-person views.

As users move through a custom world, the AI generates the environment in real time. Google claims the model simulates physics and interactions to make exploration feel more realistic. Users can also remix existing worlds available on the platform by modifying them with new prompts.

There are some limitations. Each generated world can only be explored for up to 60 seconds. Google also cautions that creations may not always look completely true to life, and users may experience higher latency while controlling characters.

Why Google Released Project Genie

At the moment, Project Genie is available only to Google AI Ultra subscribers in the US. According to Google DeepMind researcher Diego Rivas, the tool has been released primarily as an experiment to understand how people use it.

ALSO READ: iPhone 17 Demand Drives Apple’s Biggest Quarter Yet

“It’s really for us to actually learn about new use cases that we hadn’t thought about.”

Google believes that insights from Project Genie could eventually help train systems such as robots to better navigate the real world.