Google DeepMind introduced Genie 4 on Tuesday. This AI tool creates worlds on a computer. People can move through them and make changes as they go. Google says the worlds look more real than those in the earlier version, Genie 3.
Users can type simple instructions to change the weather or add people and objects. They can also change the view. Genie 4 makes sounds for the world, including music, background noise, and people talking. Another Google AI tool, Gemini 3.8 Flash, helps plan steps and check what happens.
Genie 4 can keep a scene consistent for about two minutes. Genie 3 could do this for about one minute. Google says the new tool could help people plan films, try game ideas, and create practice situations for robots.
But longer scenes can have problems. Objects can move to the wrong place or act in strange ways. Sounds can be wrong, and words inside the world can be hard to read. Only selected testers have access now. Google plans to offer it to Google AI Ultra users through Project Genie and to selected partners through other Google tools.
