Google DeepMind Releases Genie 4 World Model with Real-Time Audio and Interactive Capabilities
Google DeepMind has launched a research preview of Genie 4, its latest generative world model. The model significantly advances previous iterations by offering photorealistic environments, real-time audio generation (including ambient noise and dialogue), and the ability for users to steer simulations via text prompts mid-session. It features improved physics, more human-like characters, and a doubled scene consistency limit of up to two minutes. The release highlights an intensifying competition in the world model sector, following recent moves by Runway and industry consolidation involving AMD, Amazon, and Rivian. Google is positioning Genie 4 as a foundational tool for gaming, interactive learning, and robotics simulation. By integrating Gemini models to act as 'directors' for visual generation, Google aims to create hybrid systems capable of complex task planning and rapid creative prototyping.
Tickers: $AMD $AMZN $GOOGL $RIVN
Topics: Generative AI Product Launch Robotics Simulation World Models
Read the full article at TechCrunch
Part of the story Google Announces Genie 4 AI World Model with Enhanced Realism and Audio Generation (3 articles)