Create Your Own 3D World with Google Project Genie

Google's Era of World Creation with AI

Tech giant Google is making a new phase in its artificial intelligence research publicly available. The 'Genie 3' model, introduced by the DeepMind team last summer, is now being offered to users outside of Google under the name 'Project Genie'. This model is described as a 'world model' that can create interactive 3D virtual environments based on user descriptions.

How Does Project Genie Work?

The project offers three main interaction modes: World Sketch Creation, Exploration, and Remix. In the first stage, Google's 'Nano Banana Pro' model creates the source image that Genie 3 will use. The user can define their character, choose the camera angle (first-person, third-person, or isometric), and determine how they want to explore the world. The system presents a sketch before generating the final world, allowing the user to make changes. It is also possible to write your own text commands for worlds created by others.

However, it is emphasized that Genie 3 is not a game engine. Although its outputs may look game-like and it can simulate physical interactions, it does not have traditional game mechanics. Productions are limited to 60 seconds and presentations are made at 24 frames per second and 720p resolution.

Access Conditions and Future

Users who want to try Project Genie must meet certain criteria. Users are required to subscribe to Google's $250 per month 'AI Ultra' plan, reside in the USA, and be over 18 years old. This move is considered a testing phase on the path to commercializing AI's capabilities in the creative and simulation fields. Similarly, other AI projects like OpenClaw have also come to the forefront with their usage and cost models following their rapid rise.

Experts state that although Project Genie is currently a limited beta version, it opens an important window into what the future might hold in the field of AI-generated content and virtual worlds.