Genie 4 builds on the Gemini family. According to DeepMind, Gemini Omni anchors simulations in real-world appearance and behaviour, while other Gemini models supply the reasoning needed to keep a simulation coherent and continuous. Users can set an event in motion and then steer the scene with natural language, changing the weather, moving the camera or introducing new characters, with prompts adapted to match the surrounding scene. Between user inputs, Gemini also monitors and steers the scene to keep the simulated world progressing cohesively.

DeepMind frames world simulation as a step towards AGI, arguing it could enable safe training environments for robots and agents as well as creative tools in fields such as filmmaking, gaming and journalism. Those are the company's own stated ambitions rather than demonstrated outcomes.

For now, access is limited. Genie 4 is rolling out as a research preview to trusted testers, with Google AI Ultra subscribers getting it through Project Genie and early access partners via the Gemini API and AI Studio. A waitlist for Project Genie is open.