Google Launches Genie 4 With Audio, Real-Time Edits — SkimNews
Get the Tech newsletter
Daily tech — startups, AI labs, chips, the launches that shape the next decade. Free.
- Google launched Genie 4, an updated world model that generates interactive AI-simulated environments with improved realism and longer duration of up to two minutes, doubling the length of its predecessor.
- Google integrated Gemini 3.8 Flash into Genie 4 to power its simulations, enabling faster processing and more responsive generation of dynamic virtual experiences.
- Google added audio generation to Genie 4, allowing the model to produce soundscapes and effects that accompany its generated 3D worlds, a new capability absent in Genie 3.
- Google introduced real-time steering via natural language prompts, letting users modify ongoing simulations without restarting, enhancing interactivity for creative workflows.
- Google is releasing Genie 4 first as a research preview to trusted testers, with plans to roll it out to Google AI Ultra subscribers through a waitlist.
Why it matters: Developers and creators gain access to a more dynamic, audio-capable simulation tool that can accelerate prototyping in gaming and robotics, but the two-minute limit and shifting object permanence constrain long-form use, making near-term adoption dependent on how well Google stabilizes extended runs.
Ask SkimNews
