Google DeepMind Releases Genie 4 World Model in Research Preview with Audio and Real-Time Steering
· GlobeWire Daily Tech Desk · Source: techcrunch.com

Image: Rebecca Bellan · Photo source · Resized and letterboxed to 16:9; no AI generation
Google DeepMind has released its newest world model, Genie 4, in a research preview. According to reports from the company briefing, the model generates interactive and photorealistic worlds, introduces real-time audio generation, and allows users to make free-form edits during simulation.
Key Features of Genie 4
- Photorealism and Audio: The model creates synchronized audio, including music, ambient sounds, and interactive dialogue, alongside improved physics and humanlike characters.
- Real-Time Steering: Users can type prompts mid-simulation to alter scenes, change camera angles, modify the weather, or add objects in real time.
- Extended Consistency: Genie 4 can maintain a consistent scene for up to two minutes, doubling the limit of its predecessor, Genie 3.
- Gemini Integration: Google’s Gemini model acts as a director, planning steps and checking frames while Genie generates visuals. In a research experiment with Gemini Robotics, a Gemini 4 Argon model broke down tasks that Genie 4 simulated in real time.
Industry Context and Potential Applications
Google's release arrives amid broader consolidation in the world model sector. Competitors and related developments include Runway's GWM Worlds 2, World Labs' planned acquisition by AMD, Amazon's talks to buy Decart, and Rivian's acquisition of Third Dimension.
While pitched commercial use cases include gaming, interactive learning, and robotics, trusted testers are exploring film prototyping. Engineers noted that the tool could assist in the ideation and storyboarding phases.
Limitations and Availability
- Current Gaps: Google DeepMind acknowledged that details and objects can drift during longer runs, and challenges remain regarding audio accuracy and in-world text legibility.
- Access Status: Genie 4 is currently available as a research preview for trusted testers. It is scheduled to roll out in the coming weeks to Google AI Ultra subscribers in Project Genie, as well as early access partners of the Gemini API and Google AI Studio.
Readers can verify official details and future rollout announcements directly through Google DeepMind publications and developer channels.
AI-assisted reporting; source claims are attributed and have not been independently verified.



