Google DeepMind has announced Genie 4, a new world model that generates interactive simulations in real time with greater visual realism. The company says the model is natively multimodal, producing music, ambient noise and interactive dialogue alongside the visual scene. Genie 4 builds on the Gemini family: Gemini Omni anchors the simulated environments in real-world appearance and behaviour, while other Gemini models provide the reasoning needed to keep a scene coherent over time. Users can change a running simulation with natural language, such as altering the weather, moving the camera, or adding a person or object. Gemini adapts those prompts to match the surrounding scene and continues to steer the world between inputs. DeepMind presents the work as part of building AI that understands and simulates the world, with potential uses including training environments for robots and AI agents and new interactive experiences. The model is not yet broadly available. It is rolling out as a research preview for trusted testers, starting with Google AI Ultra subscribers in Project Genie and early-access partners of the Gemini API and Google AI Studio, and a waitlist is open. The announcement does not include pricing or independent benchmark results, and its demonstration clips were produced by the company. For Australian businesses and professionals, nothing changes in practice yet, since there is no generally available product or price to act on. Robotics, game development and simulation teams may want to join the waitlist and read the preview terms before testing. Any trial that uses personal information or customer footage would bring obligations under the Privacy Act 1988 into play, so those inputs should be avoided until the data handling is clear.