Project Genie has officially launched
As a research prototype for Google AI Ultra subscribers, allowing users to “hallucinate” playable 2D and 3D worlds. Powered by the Genie 3 world model and Nano Banana Pro, the tool transforms a single image or text prompt into a navigable 60-second environment. Sundar Pichai described Project Genie as “out of this world” after sharing a clip of an astronaut exploring a space station generated entirely by the AI’s physics engine. Basically, Project Genie is an experimental AI tool from Google DeepMind that lets you create and play inside your own 3D virtual worlds just by describing them with text or using an image
- Generated worlds simulate real-world physics, allowing for first or third-person traversal.
- “World Sketching” feature allows users to remix existing environments in real-time.
- Limited to Google AI Ultra (U.S. only for now) to manage high computational latency.
- Aims to eventually train robots by simulating billions of unique physical scenarios.
👉 Why this matters: We are moving from AI that writes stories to AI that simulates reality, fundamentally changing how games and robots are built.
Note: 
Written and summarized by our editorial team using human review & a bit of AI assistance. Edited & Approved by Debraj Paul, Founder of ArticoliNews Media-Tech