[Google Equips Genie 4 with Gemini "Director": Generating Worlds and Planning Stories Simultaneously] Beating AI News: Google DeepMind has released its next-generation world model, Genie 4. Inputting text generates explorable virtual worlds, movement is controlled by the keyboard, and weather and characters can be added at any time. It can simulate collisions, aerodynamics, and character movements, and for the first time natively generates ambient sounds, music, and real-time dialogue. Characters not only speak but also react to the scene and user input. Genie 4 leverages Gemini Omni to enhance the realism of visuals and real-world behavior, while Gemini 3.8 Flash acts as the real-time director. The latter continuously observes the scene, breaks down tasks into steps, checks whether actions are completed, and then decides the next step. Google also demonstrated Gemini 4 Argon planning a humanoid robot to play checkers, with Genie 4 simulating the actions in real time. A single simulation lasts up to 2 minutes, twice the duration of the previous Project Genie experience. However, even after running for a long time, the position and appearance of objects may still change. In the official demo, the motorcyclist suddenly appeared wearing an extra helmet in both front and rear camera shots. Genie 4 is currently only available to a limited number of invited testers, and will later be gradually rolled out to Google AI Ultra, the Gemini API, and AI Studio. Google hopes to use it to create game prototypes, preview movie scenes, and, more importantly, allow robots and AI agents to practice complex tasks in a virtual world. [Original Link]