Google DeepMind announced Genie 4 on October 27 as a research preview for trusted testers. The world model generates interactive environments with native music, ambient audio, and dialogue, and accepts text instructions that change a scene while it is running. Google says access will expand to Google AI Ultra subscribers through Project Genie and early-access partners in the Gemini API and Google AI Studio; those wider releases are still forthcoming.
Google describes uses ranging from film previsualization and game prototypes to training environments for robots and AI agents. The preview remains experimental: the company acknowledges drifting visual details and object permanence, along with unresolved issues in physics, spatial accuracy, legible text, audio accuracy, and long-term consistency.
Source · deepmind.google ↗
