● Covered by 4 sources · 4 reportsMedium impact4 neutral
Google DeepMind Releases Genie 4 World Model with Enhanced Realism and Audio Generation
Key points
- Genie 4 generates interactive virtual experiences with enhanced realism.
- The model can create audio, including music, ambient noises, and dialogue.
- Experiences can last up to two minutes, double the previous version.
- Users can update simulations in real-time with natural language prompts.
- Genie 4 uses Gemini 3.8 Flash to guide its simulations.
Genie 4 Enhances Virtual World Generation
Google DeepMind has released Genie 4, its latest AI world model designed to generate interactive virtual experiences. This new iteration offers more realistic worlds and allows for real-time updates to an experience using natural language prompts. The model also now includes the capability to generate audio, such as music, ambient noises, and interactive dialogue, to accompany its simulated environments.
Technical Improvements and Simulation Duration
Genie 4 utilizes Gemini 3.8 Flash to guide its simulations, contributing to improved consistency and the ability to change parameters mid-stream. The model can create experiences that are approximately two minutes long, an increase from the one-minute duration of the previous Genie 3 model. While the simulations are more realistic, Google notes that details and object permanence can shift over time as simulations run longer.
Applications and Research Preview
Google is positioning Genie 4 as a tool for various applications, including pre-visualization for filmmakers, rapid prototyping of game ideas, and creating training scenarios for robots. It also provides a way for developers to generate test environments on demand for AI agents to practice multi-step tasks. Genie 4 is currently available as a "research preview" to "trusted testers."
Context in the World Model Landscape
The release of Genie 4 follows Google's Genie 3 model and comes amidst increasing activity in the world model industry. Other companies, such as AI video startup Runway with its GWM Worlds 2, are also advancing in this field. The industry is seeing consolidation, with acquisitions like Fei-Fei Li’s World Labs by AMD and Amazon's talks to acquire Decart.
✨ This summary was generated by AI from the outlets' reporting listed below. It is not independently verified and may contain errors — check the original sources. How BrevFeed works →
The daily brief
One email each morning: the day's tech stories, clustered across outlets and summarized. No account needed.
One email a day. Unsubscribe in one click, any time.
Today's brief
Spend a few minutes, get the whole day. Every topic's top stories in one hands-free rundown — listen, watch, or read the transcript.
▶ Play today's brief
New every morning, and the back catalogue is archived by date.
How outlets covered it
Google DeepMind launched Genie 4, a world model that creates interactive, real-time environments with simulated physics and human motion. This allows AI agents to practice multi-step tasks in generated test environments, offering a new method for developers to evaluate agent performance.
Google DeepMind released Genie 4, its newest world model, which generates more photorealistic and interactive simulations with real-time audio and mid-simulation prompt-based editing. This release enhances Google's position in the competitive world model industry, which is seeing consolidation and advancements from other companies.
Google DeepMind released Genie 4, an updated world model that generates virtual environments and experiences frame by frame in real time. Genie 4 uses Gemini 3.8 Flash to guide simulations, improving consistency and allowing midstream parameter changes, which DeepMind states is a step towards artificial general intelligence.
Google has released Genie 4, an AI world model that generates interactive experiences with increased realism and the ability to create audio. The model uses Gemini 3.8 Flash and extends experience duration to two minutes, offering potential applications in filmmaking, game prototyping, and robotics training.