EXPLORINEER
en

DE — heise online

Google Genie 4: Interactive AI Worlds Now Available to Subscribers

AI2 min read

Google is opening its new world model, Genie 4, to paying Google AI Ultra subscribers through a web app called Project Genie. It can make interactive 3D worlds that users explore with a keyboard and change by typing.

What happened

Google is rolling out access to Genie 4 for subscribers of Google AI Ultra, step by step. The service is also available in Germany. Users of the cheaper Google AI Pro plan can join a waiting list. Daily limits apply. Genie 4 creates interactive 3D places from text or pictures. Users can explore them with the keyboard and change them while they play by typing new instructions. Compared with Genie 3, the new model adds sound and dialogue, sharper graphics, more natural movement, and more believable physics.

A helper agent and a community gallery

Project Genie now has an AI agent that helps people design their worlds. Users describe their ideas in a chat, and the agent uses several Google models to make starting images and matching background music. While users explore, Gemini can check what is on the screen, suggest next steps, or trigger events in the world. A new community gallery lets users share their worlds, explore others' creations, and change things like the setting, the music, and the 'rules of the game'.

Key facts

  • Genie 4 supports about one minute of continuous interaction.
  • Resolution is 720p, with up to 24 frames per second and synced sound, according to the newly published model card.
  • Genie 3 stays available in Project Genie. Its sessions now last up to 3 minutes, up from 1 minute.
  • Users can link several Genie 3 worlds into one longer trip, and they can start from their own 360-degree panorama photos.

Still experimental

Google says this technology is still experimental. Camera controls can respond with a delay. Physical actions are not always simulated correctly. The simulation can also struggle to stay consistent over longer periods. Genie 4 was first shown at the end of October.

Why it matters

This shows how AI may soon let people build and play in their own interactive worlds, not just watch videos. The story also points to a split between paying tiers, since the cheaper plan has to wait. Google's model also drew attention in the gaming market, where related reports said gaming stocks fell after people tried it.

Glossary

Genie 4
Google's AI world model that turns text or images into interactive 3D environments.
Project Genie
The web app where people can try Google's Genie world models.
Google AI Ultra
A paid Google AI subscription with the highest access level, including Genie 4.
Google AI Pro
A cheaper paid Google AI subscription that currently uses a waiting list for Genie 4.
Gemini
Google's AI model family that can read screen content and suggest or trigger actions.

Ad

Add 150 words. Learn 10 a day.

Dicter word books send only the words you learn to your review notes.

AI전자신문

Lablup Posts 48% Year-on-Year Revenue Growth for Nine Months, First Earnings Report Since Listing

Lablup reported cumulative consolidated revenue of 6.99 billion won and operating profit of 900 million won for the first three quarters, up 48.2% and 36.5% year on year, driven by corporate demand for AI infrastructure. It posted a cumulative net loss of 2.4 billion won due to a 3.49 billion won derivative liability valuation loss tied to the conversion of RCPS shares, but returned to a net profit of 360 million won in the third quarter as that accounting effect expired.

AIZDNet Korea

Maum AI Signs Physical AI Industry-Academia MOU with Hanyang University ERICA and Doosan Robotics

Maum AI signed a three-way MOU with Hanyang University ERICA and Doosan Robotics at the "AI Symposia" held on the 18th at Hanyang University ERICA to build cooperation on physical AI talent development and industry-academia collaboration. The partnership is based on the AX Graduate School project, a government-backed AI talent program run by the Ministry of Science and ICT, with cooperation planned in joint curricula, industry projects, internships, joint R&D and demonstrations, and government R&D planning. Maum AI will contribute its generative and multimodal AI and AI platform capabilities to the effort.

AIheise online

Gemini 4 Neon Challenges Google's Flagship Model

Google DeepMind has released Gemini 4 Neon, the fourth model in the Gemini 4 family within seven weeks, designed for efficiency while matching and slightly outperforming flagship model Carbon on some benchmarks. Neon is slated to become cheaper per token than Carbon from December, when Carbon's introductory prices expire. Google positions it for complex programming, analytical reasoning and enterprise use, filling the gap between Carbon and the everyday model Hydrogen.

AITechNode

Tencent says Hunyuan Hy4 flash surpasses 50 trillion tokens in first-day usage

Tencent's Hunyuan said Hy4 flash processed more than 50 trillion tokens on its first day after its November 16 launch, the highest first-day usage in the Hunyuan series. The model ranked first on OpenRouter's daily LLM leaderboard on November 17, and Tencent says it is priced at 0.5 yuan per million input tokens and 2 yuan per million output tokens.