Skip to content

CreativeGoogle DeepMind

Google DeepMind opens Genie 4 in Project Genie to Google AI Ultra subscribers

Genie 4 at a glance

Google DeepMind opened Genie 4 to the public on November 18, 2026, inside a new Project Genie experience for Google AI Ultra subscribers: the interactive world model now takes mid-session prompts and real-time native audio, and Genie 3 sessions run up to 3 minutes.

Ask your AI about this story

Ask what changed in Genie 4, Project Genie and Genie 3 and what to check before using it in your business. Weigh quality, cost per output, licensing and commercial-use terms, and how well it keeps a brand's style.

Sources checked Primary source (opens in a new tab)

What changed with Genie 4

Google DeepMind opened Genie 4, its interactive world model, to eligible Google AI Ultra subscribers on November 18, 2026, inside a new Project Genie experience that runs in the browser. Google DeepMind describes Genie 4 as its most realistic and interactive world model yet: while you explore a generated world, you can type a new prompt mid-session to shift the weather, drop in a character or redirect the action, and the world answers with real-time native audio, from ambient soundscapes to character dialogue.

The release also upgrades Genie 3, which stays in Project Genie's model picker for longer runs: sessions now last up to 3 minutes per generation, up from 60 seconds, worlds can be chained together for longer journeys, new worlds can be seeded from 360-degree panorama images, and explorers can trigger in-world events with natural-language prompts. A new Project Genie agent builds worlds conversationally, orchestrating Google's multimodal models for the visuals and the background score, and a community gallery lets subscribers publish worlds for others to explore and remix.

Google AI Pro members can join a waitlist for Project Genie access on the same page; Google announced no separate price or country list beyond the AI Ultra subscription. Google says Genie 3 and Genie 4 remain frontier research, with camera latency, intricate physical interactions, real-time audio precision and long-term scene permanence as open challenges, and that all generated video and audio carries SynthID watermarking and passes safety screening. Genie 4 was first unveiled on October 27, 2026, when access was limited to trusted testers.

Why Genie 4 matters for businesses

Why it matters: Project Genie turns world models into something a business can actually try today. A studio, trainer or marketer can prototype an explorable demo, a virtual showroom or a training simulation from a prompt and a conversation instead of commissioning a game-engine build, and can restyle it mid-session while a client watches. Access is limited to AI Ultra subscribers and the output is still research-grade, so it fits prototyping and pitching, not a production pipeline.

Put it to work: AI services

Get AI News by email

Verified AI stories, each with its official source, on Monday and Thursday mornings. Add breaking news to hear about the biggest ones as soon as we publish them.

Every story links its official source

Free, in English or Spanish

Unsubscribe with one click, any time

What to send

Sources for this story

  1. Official

Last verified: . IA Compa summarizes public announcements for business readers. We are not affiliated with the companies mentioned. Always confirm details, pricing, and availability with the official source.

  • #google-deepmind
  • #genie-4
  • #project-genie
  • #genie-3
  • #world-model
  • #interactive-worlds
  • #google-ai-ultra
  • #synthid

Want this working in your business?

We turn new AI capabilities into agents, voice receptionists, and automations you can actually use.