Google DeepMind's Genie 4 world model enters limited research preview
The AI builds interactive virtual worlds with sound and dialogue, and users steer them through plain-language instructions as they run.
By Standard Topic. AI-generated from the reporting listed under Coverage. Published . Updated .
This story is no longer being updated. It reflects reporting up to .
What happened
Google DeepMind's Genie 4 generates interactive simulated environments in real time, producing visuals, environmental sounds, music and dialogue. Users can change a running scene with natural-language prompts. Access is currently limited to trusted testers, while Google AI Ultra subscribers and early API partners are expected to get it later.
Why it matters
Users can alter scenes midway, such as changing the weather or adding a person, which expands how the tool can be used. DeepMind says the technology is still at an early research stage.
Timeline
- Oct 27, 10:50 PM ET: Google DeepMind announces Genie 4, an AI that generates virtual worlds in real time.
- Oct 28, 12:14 PM ET: Genie 4 rolls out to trusted testers, with Google AI Ultra and early API partners to follow.
- Oct 28, 12:59 PM ET: DeepMind says the technology is still in an early research stage despite impressive demonstrations.
- Oct 28, 5:18 PM ET: Google says SynthID will mark all Genie 4 output, while only trusted testers currently have access.
Coverage
- Google DeepMind unveils Genie 4: Why Sundar Pichai calls it ‘our most realistic and interactive world model' (CNBCTV18.com)
- Genie 4: Google's new AI world model generates complete scenes, including sound and dialogue (basic-tutorials.com)
- Google launches Genie 4 world model for select game, film teams (digitimes.com)
- Google’s Genie 4 builds worlds you steer with words, and its characters talk back (martincid.com)
- Google's Genie 4 simulates people who talk back in real time (ppc.land)
- ICYMI: Google DeepMind previews Genie 4 with native audio (Erin)
- Google DeepMind Unveils Genie 4: All About Real-Time World Building With Built-In Sound (NDTV)
- Trusted testers get first Genie 4 access; Ultra and API rollout is still to come (OpenTools Team)
- Google DeepMind’s Genie 4 world model adds sound and live steering (Ana Maria Constantin)
- Google DeepMind previews Genie 4 world model (Kerem Gülen)
- Google’s AI worlds can now talk back (therundown.ai)
- Google's Genie 4 Can Now Generate Sound and Modify AI Worlds in Real Time (mezha.ua)
- DeepMind unveils Genie 4 world model with sound and live text steering (Resultsense)
- Google DeepMind's Genie 4 Alters Virtual Worlds in Real Time via Commands (chosun.com)
- Google built AI agents a world to practice in. The physics still aren’t quite right. (thenewstack.io)
- Google Genie 4 Builds Living Worlds With Sound Physics and Real Time Control (techeblog.com)
- Google's New AI World Model Can Generate Human Voices, Not Just Video (gizmodo.com)
- Google Genie 4: Specs, Benchmarks, Videos and Access (kingy.ai)
- Google Gains Over 2% After Unveiling World Model Genie 4 Ahead of Earnings, Poised to Expand Revenue Streams (tradingkey.com)
- Google DeepMind unveils Genie 4 with sound and world editing (ua.news)
- Google's Genie 4 Conjures Explorable Worlds in Real Time, With a Short Memory (Around Prague)
- Google DeepMind is unveiling Genie 4, its most realistic AI world model yet (qz.com)
- Google DeepMind Unveils Genie 4 World Model With Native Audio And Real-Time Steering (NewsCord)
- Google Deepmind's Genie 4 puts a modern spin on agent designs that predate LLMs (the-decoder.com)
- Google DeepMind unveils Genie 4, the world model that once tanked gaming stocks (startupfortune.com)
- Google DeepMind Genie 4 Adds Native Audio and Live Scene Edits (sqmagazine.co.uk)
- Google DeepMind Genie 4 Released: Real-Time World Simulation & Native Audio (Nayan)
- Google DeepMind Rolls Out Genie 4 World Model in Research Preview (unite.ai)
- Google’s next Genie AI model generates more realistic experiences (theverge.com)
More from Technology & AI
- China puts humanoid robots to work serving food and sorting parcels
- Newsom proposes sweeping AI safety rules for Washington
- Google releases Gemini 4 Neon AI model with million-token context
- Microsoft and Intel widen their cloud chip partnership
- Google DeepMind rolls out Genie 4 to select subscribers
- AI helped AMD shorten chip design work, OpenAI says