Google DeepMind previews Genie 4 with native audio and interactive world control
Discovery source: Google DeepMind · Source date:
On October 27, 2026, Google DeepMind said it is previewing Genie 4, a new world model that the company says improves real-time interactive simulation and is natively multimodal, generating music, ambient sounds, and interactive dialogue alongside visuals. DeepMind says the model is starting as a research preview for trusted testers, with access beginning for Google AI Ultra subscribers in Project Genie and early-access partners in the Gemini API and Google AI Studio. In the announcement, Google describes Genie 4 as capable of real-time scene edits, more realistic physics and human motion, and native audio generation. These are Google’s claims from a preview announcement; the post does not provide independent evaluation or measured benchmarks.
Check the sources
This is a NextGen-AI briefing, not a reproduction of the discovery source. A newsletter may cover several developments; consult the evidence below for this briefing’s specific subject. The source date refers to the discovery entry, not necessarily the announcement date.