Google DeepMind unveils Genie 4, a real-time world model that generates audio natively
Google DeepMind announced Genie 4, its newest world model, as a limited research preview for trusted testers on 27 October 2026. The system generates simulated worlds frame by frame in real time, producing ambient sound, music and character dialogue alongside the visuals, and it is steered through natural-language prompts. Gemini 3.8 Flash acts as a guiding agent that breaks scenarios into steps and adjusts when output drifts from intent. Genie 4 improves on Genie 3 with more realistic physics and human movement, and all outputs carry SynthID watermarking. Access is planned via Project Genie for Google AI Ultra subscribers and through the Gemini API and Google AI Studio for early-access partners, with no prices or dates yet published, and visual drift and approximate physics remain acknowledged limitations.
This is an AI-generated summary. ShortSingh links to the original source for the complete article.

Discussion (0)
Log in to join the discussion and vote.
Log in