
What happened
Google announced Genie 4, a world model that generates interactive experiences about two minutes long, up from one minute on Genie 3, with four main updates including real-time natural-language editing and audio.
Why it matters
Google is pitching Genie 4 for pre-visualization, game prototyping, and robot training, but warns that longer simulations can shift details and object permanence over time.
What to watch
Genie 4 is launching first as a "research preview" to "trusted testers" and is "coming soon" to Google AI Ultra subscribers, with a waitlist now open for Project Genie.
WHO IT HITSFilmmakers doing pre-visualization, game designers prototyping ideas, and robotics teams building training scenarios are the roles Google names as potential users, though the model is initially limited to trusted testers.
Summaries like this, every morning.
Genie 4 builds directly on Genie 3, the model behind Project Genie, which The Verge's Jay Peters previously used to generate what he describes as bad Nintendo knockoffs. The jump from one-minute to about two-minute experiences is the headline spec change, but Google is also adding real-time editing through natural language and audio generation, plus Gemini 3.8 Flash as the underlying model for simulations. Google is positioning the tool for pre-visualization by filmmakers, game idea prototyping, and robot training scenarios. The company itself cautions that as simulations run longer, details and object permanence can shift over time, a limitation worth noting for anyone planning to rely on the output. Access is gated for now: a research preview for trusted testers, a waitlist for Project Genie, and a "coming soon" status for Google AI Ultra subscribers.
Pick your industry and the AI tools you use, and get news related to your work every day.
Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →
Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.
Tiny Desk AI now sells one-time video packs starting at $9.99, using one photo per seat to cast you and friend…

Bona Film Group's Sanxingdui: Future Past, China's first fully AI-generated feature to reach cinemas, took in…

NTTドコモ announced it has developed a Physical AI連携基盤 (a linkage platform for Physical AI), positioned around a…

The first seven-minute episode of Elon Musk's AI-generated 'The Odyssey', made with Grok Imagine in collaborat…

Percepto launched Axion on October 27, 2026, a SaaS platform applying more than 160 AI models trained on over…

Crunchbase data shows global robotics venture funding reached $9.9 billion in Q3, more than double the $4.3 bi…
