Google Launches Genie 4 with Native Audio and Real‑Time World Editing
google
| Source: TechRepublic | Original article
Google has launched Genie 4, featuring native audio support and real‑time world editing, with details on US pricing and pilot considerations.
Google has rolled out Genie 4, the next iteration of its AI‑driven world‑model platform, adding native audio generation and real‑time scene editing capabilities. The DeepMind team says users can now steer photorealistic environments with natural‑language commands – moving the camera, altering weather, inserting characters or objects – while the system produces ambient sounds, music and interactive dialogue on the fly. Access is being opened to U.S. customers, with pricing and usage limits disclosed for pilot projects, though the exact figures have not been released.
The upgrade matters because it pushes generative AI beyond static image or video synthesis into fully interactive, multisensory experiences. By coupling visual realism with synchronized audio, Genie 4 opens new possibilities for entertainment, gaming, virtual production and training simulations, where creators can iterate on scenes instantly rather than rebuilding assets. The real‑time edit feature also lowers the barrier for non‑technical teams to prototype immersive content, potentially accelerating adoption across media and enterprise use cases.
As we reported on 29 October 2026, Google’s earlier Genie 4 demo already showcased photorealistic worlds with sound. This latest announcement expands that foundation with live manipulation and a clearer commercial rollout. The next steps to watch include the rollout of the pricing tiers, the definition of token‑based limits for audio‑visual generation, and how early adopters integrate Genie 4 into existing pipelines. Competitors such as Dream’s construction‑focused world models and other multimodal platforms will likely respond with their own real‑time, audio‑enabled offerings, shaping the next wave of immersive AI tools.
Sources
Back to AIPULSEN