Google has launched a new AI experiment called Project Genie, a tool that allows users to create their own interactive virtual world. Project Genie comes from the Google DeepMind research lab and is now available to Google AI Ultra subscribers in the United States (users must be over 18).
If you’ve signed up for AI Ultra — the AI subscription plan costs $249.99 — you can start building worlds right away by visiting Google Labs and navigating to Project Genie.
What does Project Genie do?
A video released along with Project Genie shows exactly how users can create custom worlds. Most importantly, Project Genie doesn’t just create a virtual environment; it also allows the user to create a character to explore and interact with the world. Users can also use the prompts to create their own mini-games, using their keyboard’s arrow keys to control the character. (One reporter at The Verge quickly realized that he could use Project Genie to make Zelda and Super Mario score.)
Google says Project Genie can render the world at 720p resolution and 24 frames per second.
Considering how much time and work goes into creating video games, it’s surprising that Project Genie can create detailed, interactive environments using AI. That said, Project Genie is primarily a research project, and while it can be fun to play with, it doesn’t really do that. do anything else. (Read on to learn about upcoming apps.)
Mashable Light Speed
The created worlds have realistic physics, and objects in the world react when the user interacts with them. Users can rearrange and remix their virtual worlds to their heart’s content.
On the left:
Create a virtual world from scratch.
Credit: Google / YouTube
Right:
Users can also change the world as they wish.
Credit: Google / YouTube
Project Genie is powered by a ‘world model’ for testing.
Project Genie is powered by Genie 3, a powerful “world model”. (Google says Nano Banana and Gemini are also used to generate worlds.) A world model is an AI program that can generate a virtual world with text, images, and other inputs. Google actually teased Genie 3 in August 2025, calling it “an important stepping stone on the road to AGI.”
The new Project Genie landing page elaborates: “Genie 3 represents a huge leap in capabilities – allowing agents to predict how the world is evolving, and how their actions affect it. Genie 3 makes it possible to explore an unlimited range of virtual environments. This is an important step on the path to AGI – allowing AI agents to think, solve real-world problems, and solve problems.”
Artificial general intelligence, or AGI, is the term for a hypothetical AI tool that can perform as many tasks as (or more) than a normal human. That means it can do its job with little or no supervision. To achieve true AGI, AI companies need to build models that can make sense of the environment and understand how to interact with it.
Examples of virtual worlds created by Genie 3.
Credit: Google
World models are an emerging frontier in AI research, and companies creating video models have attracted significant investment over the past year. AI company WorldLabs recently raised $230 million in funding, while video modeling maker Luma AI raised $900 million.
Besides AGI, what are the practical uses of world models? As one example, car companies can build a world to test autonomous vehicles. There are also potential applications for video game development and education.
Also, AI Ultra subscribers in the United States can start using Project Genie right away.
A virtual world created by Project Genie.
Credit: Google / YouTube
A virtual world created by Project Genie.
Credit: Google / YouTube
A virtual world created by Project Genie.
Credit: Google / YouTube
Articles
Artificial Intelligence Google
