Skip to content

Google opens Genie 4 world model to AI Ultra subscribers

Short answer

Google DeepMind has released its Genie 4 world model inside Project Genie to eligible Google AI Ultra subscribers. Users can edit a running world by typing prompts and get generated audio. Genie 3 also gained longer sessions. The source mentions no API or business plan, so operational impact is minimal for now.

What this means for operators

If you run sales, support or operations at a company of 10-200 people, there is nothing to schedule here. The source describes a browser experience for individual Google AI Ultra subscribers and says nothing about an API, team seats, integrations or business licensing, so there is no way to connect it to a CRM, helpdesk or internal process. Whether such tools could ever serve training or demo content is not addressed by Google and would be speculation. The one practical check: if someone on your team already holds an Ultra subscription, treat Project Genie as an experiment and keep it away from customer data until Google documents terms for work use. Revisit when an API or business plan appears.

Google DeepMind announced on Nov 18, 2026 that a new Project Genie experience is rolling out to eligible Google AI Ultra subscribers. It gives them first access to the Genie 4 world model, upgrades to Genie 3, a creative agent and a community gallery. Google AI Pro members can join a waitlist on the same page.

What Genie 4 does

Genie 4 is described as the company's most realistic and interactive world model so far. While moving through a generated world, a user can type a new prompt and the environment changes on the spot: different weather, an added character, a shifted storyline. Genie 4 also produces audio in real time, from background soundscapes to character dialogue, and the sound adapts as the scene changes.

Genie 3 updates

Genie 3 stays available in the model picker for people who want longer runs or panorama-based spaces. Its changes:

  • A single generation can now last up to 3 minutes, up from 60 seconds.
  • Several worlds can be linked into one longer journey.
  • New worlds can start from 360-degree panorama images.
  • The environment can be altered mid-exploration with natural-language prompts.

The agent and the gallery

The Project Genie agent lets a user describe a world in conversation. It coordinates Google's multimodal models to produce the opening visuals and a matching music score. Gemini models can also join a session: they watch the screen, suggest where to go next and trigger in-world events as the journey unfolds.

The new gallery holds curated worlds and community creations. Users can publish their own worlds, and others can remix them by changing the structure, soundtrack or rules. Examples include a photo-based explorable space, a prehistoric habitat built on panoramas and a mystery set in a gothic manor.

Safety and limits

Google says all video and audio from Project Genie carries SynthID watermarking, and that inputs and outputs pass through safety screening. It also calls Genie 3 and Genie 4 frontier research and lists open problems: camera latency, complex physical interactions, real-time audio precision and long-term scene permanence.

What is not in the announcement

The post gives no pricing, no developer access and no business or team plan. Everything it describes happens in a browser for individual subscribers, so any workplace use is outside what Google has said so far.

Source: Google DeepMind

The tools in this story

  • Gemini - price, who uses it, alternatives
Next step

Visibility Analyzer

An AI news item will not tell you how assistants see your own site. The Visibility Analyzer checks that on your live site: which answers cite you, which pages an engine cannot retrieve, and what to fix first. Free to run.

Run a free visibility audit

Free to run. No card.

Fee
Free
Length
One run, minutes

Free tier: two analyses a day, no card required.