Krux
February 23, 2026
Google Launches Project Genie for AI-Generated Interactive Worlds
Published: February 23, 2026 at 12:27 AM
Updated: February 23, 2026 at 12:27 AM
100-word summary
Google DeepMind released Project Genie, an experimental prototype now available to Google AI Ultra subscribers in the U.S. Powered by Genie 3, the tool lets users create and explore interactive worlds from text or image prompts with real-time path generation and dynamic simulation. Users can sketch worlds with adjustable first- or third-person camera views, remix existing environments, and export exploration videos. Known limitations include 60-second generation caps, occasional physics mismatches, and character control latency. The release signals Google's push toward practical world-model AI, expanding immersive creation tools beyond research teams into subscriber hands today.
What happened
Google DeepMind released Project Genie, an experimental prototype now available to Google AI Ultra subscribers in the U.S. Powered by Genie 3, the tool lets users create and explore interactive worlds from text or image prompts with real-time path generation and dynamic simulation. Users can sketch worlds with adjustable first- or third-person camera views, remix existing environments, and export exploration videos. Known limitations include 60-second generation caps, occasional physics mismatches, and character control latency.
Why it matters
The release signals Google's push toward practical world-model AI, expanding immersive creation tools beyond research teams into subscriber hands today.