The Genie 4 world model delivers greater realism and includes audio generation capabilities to support its interactive environments.
Google announced the Genie 4 world model today for generating interactive experiences using AI. The model boasts four major updates: it allows for more realistic worlds, permits users to update experiences in real time using natural language prompts, generates audio, and runs on the Gemini 3.8 Flash architecture to assist with simulations. Furthermore, it can create experiences lasting approximately two minutes, compared to the one-minute duration of the prior Genie 3 model.
Google is positioning Genie 4 as a tool beneficial for tasks such as film pre-visualization, game idea prototyping, and creating training scenarios for robots. However, the company warns that as simulations run longer, details and object permanence can shift over time.
The author mentions they have not had the opportunity to test Genie 4 personally, as it is launching as a "research preview" for "trusted testers," preventing a direct comparison to the Genie 3-powered Project Genie, which the author previously used to generate Nintendo knockoffs. Google has provided examples demonstrating the potential of the new model.
Genie 4 will be "coming soon" for Google AI Ultra subscribers, according to Google. The company has also opened a waitlist where individuals can apply to try the model within Project Genie.
The Editorial Staff at AIChief is a team of professional content writers with extensive experience in AI and marketing. Founded in 2025, AIChief has quickly grown into the largest free AI resource hub in the industry.