Economy 51
Imagine creating a realistic world with a simple text prompt, changing the weather or camera angle instantly and hearing sounds that match every action. Google DeepMind is aiming to make this possible with Genie 4, a new AI model designed to create realistic, interactive simulations in real time. Announced on October 27, Genie 4 combines realistic visuals, real-world physics, reasoning and audio generation to make virtual environments more lifelike.
In a post on X , Google CEO Sundar Pichai described Genie 4 as the company’s "most realistic and interactive world model". Google DeepMind has started rolling out Genie 4 as a research preview to trusted testers. The model is expected to become available to Google AI Ultra subscribers through Project Genie, as well as to early-access partners through the Gemini API and Google AI Studio.
Excited to share Genie 4 today our most realistic and interactive world model: - steer dynamic environments in real time, shaping events, camera angles, and actions - generate native audio matched to the world, from realistic background sounds to characters who can converse in real time! - simulate photorealistic scenes and real-world physics - create more coherent multi-step events through reasoning from Gemini We've started to roll out as a research preview to trusted testers. Coming soon to Google AI Ultra subscribers in Project Genie and early access partners via the Gemini API and AI Studio. — Sundar Pichai (@sundarpichai) October 27, 2026 Create and control virtual worlds in real time Genie 4 allows users to set events in motion and steer scenes in real time using natural language, including moving the camera, changing the weather or introducing new characters. Genie 4 also draws on other Gemini models for the complex reasoning needed to maintain a simulation’s coherence and continuity.
Realistic visuals with built-in audio One of Genie 4's key features is its ability to generate audio. This includes environmental sounds, background music and interactive dialogue between characters. The model also aims to create photorealistic scenes and simulate real-world physics.
How Gemini models power Genie 4 Genie 4 builds on capabilities from Google' s Gemini family of models. Gemini Omni helps ground simulations in realistic appearances and behaviours, while other Gemini models provide reasoning capabilities to maintain continuity and coherence. Pichai highlighted how this reasoning can affect a simulation in real time.
In one demonstration, a user prompts a traffic light to turn green. The model then reasons through the next sequence of events: the motorcycle engine revs and the bike accelerates along a coastal road. The demonstration illustrates how a simple instruction can trigger a connected series of actions within a virtual environment.
Potential uses across industries Google DeepMind sees world models such as Genie 4 as an important step towards artificial general intelligence (AGI). Such models could help create controlled training environments for robots and AI agents. The technology could also support creative work across filmmaking, journalism and gaming, as well as in manufacturing.
Tags Home / Technology News / Google DeepMind unveils Genie 4: Why Sundar Pichai calls it ‘our most realistic and interactive world model'
Source: CNBC TV18
