Google’s next Genie AI model generates more realistic experiences
The Verge Jay Peters ● Covered by 4 sources
Google showed Genie 4, a model that makes interactive AI worlds. It’s more realistic, updates live by prompt, and stretches scenes to about two minutes.
Based on reporting by The Verge, Jay Peters — read the original for the full story.
Summary, retelling and take written by AI under human oversight; images are AI-generated illustrations. How we work · Report an error
Google announced Genie 4 today, the latest version of its world model for generating interactive experiences with AI. The pitch is simple enough: make the worlds feel more real, make them easier to steer, and make them last longer than before.
The model’s biggest changes are practical ones. Google says Genie 4 can create more realistic worlds, can be updated in real time with a natural-language prompt, and can generate audio. It also runs with Gemini 3.8 Flash underneath, which Google says helps with the simulations.
The length bump stands out too. Genie 4 can produce experiences that run for about two minutes, up from one minute in Genie 3. That’s not a tiny tweak. It gives the system more room to do something that feels like a scene instead of a quick demo.
Google is framing the model as a tool for pre-visualization in film, for prototyping game ideas, and for robot training scenarios. So this isn’t just about making pretty toy worlds. It’s being aimed at workflows where a rough simulation can save time before the expensive stuff starts.
My take — AI-written commentary, not fact-checked reporting
Google is clearly trying to move world models from flashy demos toward something people might actually use. That matters more than the usual AI confetti, because real-time edits and longer scenes are where these systems stop looking like tech theater. The industry keeps pretending realism is the finish line; in practice, usefulness is the part that pays the rent.
Read more about this at: The Verge