🔍 Read the full analysis: How Google’s Genie 4 Creates Worlds You Can Move Through on ThorstenMeyerAI.com
Get the latest gadgets delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
TL;DR
Google DeepMind presented Genie 4 on October 27, describing it as a real-time world model that generates scenes as users move through them and issue text instructions. The company says it adds native audio and more realistic motion, but has not published benchmark results; access is limited to trusted testers, and scenes may drift during longer sessions.
Google DeepMind presented Genie 4 on October 27, describing an AI system that generates a scene as a user moves through it and changes it with text instructions. The announcement matters because it points beyond fixed AI-generated video toward interactive, sound-enabled environments, though Genie 4 is not yet generally available and its demonstrations were selected by the company.
In DeepMind’s examples, a user enters a scene description, navigates with keyboard or mouse controls, and types instructions such as changing the weather. The system generates what happens next rather than playing back a finished clip. DeepMind says the model can create ambient sound, music and dialogue alongside the visuals, and can respond to instructions while a scene is running.
The company also describes a separate Gemini system working with Genie 4 to interpret goals, break them into steps and monitor generated frames. DeepMind says this setup helps keep interactions coherent. Those capability descriptions come from Google’s announcement and its own demonstrations; the company did not publish benchmark numbers in the supplied materials.
Access is currently restricted to trusted testers. DeepMind says it plans to offer Genie 4 to Google AI Ultra subscribers through Project Genie and to early Gemini API partners, but it has not provided a release date or API price. Project Genie was previously opened to paying US subscribers for Genie 3, according to the source material; that earlier access does not mean Genie 4 is available to the public.
Infographic 01 · How Genie grew up
From flat sketches to worlds with sound, in under three years.
Each step as Google DeepMind described it. Session limits and consistency are different measures, so read the labels.
- Feb 2024
Genie 1Simple 2D worlds11 billion parameters, learned from unlabelled internet videos - Dec 2024
Genie 23D worlds from a single imageConsistent “for up to a minute”; real time only as a slimmed-down, lower-quality version - Aug 2025
Genie 3First real-time Genie720p at 24 frames per second, consistent “for a few minutes”, memory about one minute back - Jan 2026
Project GenieUS AI Ultra subscribers can play60 seconds per world, partly because of compute costs - May 2026Project Genie goes globalIn the $200 AI Ultra tier, for users 18 and older
- 27 Oct 2026
Genie 4Sound, live steering, more realismResearch preview for trusted testers; Ultra subscribers and API partners “coming soon”
Genie 3 vs. Genie 4
As described by Google DeepMind. Genie 4 column = vendor claims unless marked.
| Feature | Genie 3 (Aug 2025) | Genie 4 (Oct 2026) |
|---|---|---|
| Sound | Genie 3: None | Genie 4: Generated with the picture: ambient noise, music, dialogue |
| Steering | Genie 3: Keys, plus “promptable world events” | Genie 4: Free-form text changes while it runs; Gemini adapts and checks them |
| Look | Genie 3: 720p, 24 fps | Genie 4: “Unprecedented realism”; no resolution or frame rate published |
| How long | Genie 3: “A few minutes”; Project Genie capped at 60 s | Genie 4: “A few minutes” (DeepMind); up to ~2 min consistent scene reported |
| Access | Genie 3: Small preview, then Project Genie | Genie 4: Trusted testers; Ultra and API partners “coming soon” |
Sources: Genie paper, arXiv, 23 Feb 2024; Genie 2 blog, 4 Dec 2024; Genie 3 blog, 5 Aug 2025; TechCrunch, 29 Jan 2026; Google, 19 May 2026; Google DeepMind, Genie 4 blog, 27 Oct 2026; Genie 4 model page (limitations, disclaimer); TechCrunch, 27 Oct 2026. Table wording: Thorsten Meyer.
Powered by Thorsten Meyer AI
From Generated Clips to Interactive Scenes
Genie 4’s proposed shift is from generating a clip that viewers watch to generating a world that responds as someone moves through it. If that approach works reliably, it could give game developers and filmmakers a faster way to explore settings, action or visual ideas before building them with conventional tools. DeepMind reports that testers have used Genie for film previsualization and game prototypes; these are reported examples, not independently measured productivity gains.
The system could also matter to robotics research. Simulated environments can let researchers test tasks without immediately relying on a physical robot. DeepMind’s demonstration of a humanoid playing checkers was a simulation, not a real-robot trial, so it does not establish that Genie 4 can train robots effectively in real-world conditions.
Generated scenes that include sound may also make synthetic media more convincing. DeepMind says Genie 4’s video and audio outputs carry an invisible SynthID watermark. The supplied material does not explain how reliably the watermark can be detected or preserved when content is edited, so it should not be treated as proof that every copy will remain identifiable.
Products Worth Considering
Apple AirPods Pro 3 Wireless Earbuds with Active Noise Cancellation
Bose QuietComfort True Wireless Earbuds with Active Noise Cancellation, Black
Sony WH-1000XM6 Wireless Noise-Canceling Over-Ear Headphones, Blue, Bundle
As an affiliate, we earn on qualifying purchases.
How Genie Reaches This Stage
Genie is part of DeepMind’s work on world models: systems that predict how a scene changes in response to previous frames and user input. Unlike a conventional game engine, the model described by DeepMind does not rely on a stored 3D map of every object and surface. It generates subsequent views based on what it has seen and what the user does.
The series has developed from simpler world generation. The source material says the first Genie, introduced in February 2024, generated simple 2D worlds. Genie 2 followed in December 2024 with 3D scenes that could remain consistent for up to a minute, while Genie 3, presented in August 2025, added real-time interaction. DeepMind’s earlier account of Genie 3 cited 720p resolution and 24 frames per second, with consistency lasting a few minutes.
Genie 4’s stated additions include generated audio, live text-directed changes and more natural-looking people and physics. The comparisons and claims about improvement are DeepMind’s descriptions. The source material says independent benchmark figures were not published, making it difficult to measure the size of the gains against previous models or competitors.
“Objects “can shift or drift over time.””
— Google DeepMind
Limits on Stability and Access
It is not clear how Genie 4 performs across a broad range of prompts or how its consistency compares with other systems: DeepMind published no benchmark numbers in the material provided. The demonstrations were chosen by the company, so they do not show how often unusual inputs produce convincing or stable results.
DeepMind says scenes may drift and describes interaction lasting a few minutes. TechCrunch and Bloomberg, as cited by heise, reportedly put some scenes at about two minutes, but that figure is not in DeepMind’s own announcement. The source material also does not establish a fixed session limit for Genie 4.
Availability remains limited and the timetable is unresolved. The source describes robotics as a research demonstration and says the checker-playing Apollo was simulated, with no real robot involved. It is also unclear how much the system will cost to use through the Gemini API, how broad subscriber access will be, or how the SynthID watermark performs after editing or reposting.
Awaiting Public Access Details
The next expected step is access for Google AI Ultra subscribers through Project Genie and early partners using the Gemini API. DeepMind has not announced when either group will receive Genie 4, whether access will expand beyond them, or what API use will cost.
Further evidence will depend on access beyond company-selected demonstrations. Independent tests could clarify how long scenes remain consistent, how well text instructions work, and whether generated sound stays aligned with the action. Until then, Genie 4’s current capabilities are best understood as vendor-described features and demonstrations, with practical performance and availability still developing.
Key Questions
What is Google Genie 4?
Genie 4 is a world model presented by Google DeepMind. It generates scenes as users move through them and give instructions, rather than producing only a fixed video clip.
Can the public use Genie 4 now?
No. The source says access is limited to trusted testers. Google has described future access through Project Genie for AI Ultra subscribers and through the Gemini API for early partners, but has not announced a date.
Does Genie 4 generate sound?
DeepMind says it generates audio with the scene, including ambient sounds, music and dialogue. These capabilities have been described by the company; independent performance data was not provided in the source material.
How long can a Genie 4 world stay consistent?
DeepMind says Genie 4 supports a few minutes of continuous interaction and warns that objects can shift or drift. Reports of scenes lasting about two minutes are attributed to TechCrunch and Bloomberg via heise, not to DeepMind’s announcement.
Was the humanoid robot shown playing checkers real?
No. The checker-playing Apollo demonstration described in the source was a simulation; it did not involve a real robot performing the task.
Source: ThorstenMeyerAI.com
