Google Launches Genie 4 with Real-Time World Editing for AI Ultra Subscribers
Google has officially rolled out Genie 4, the latest iteration of its Project Genie interactive world generation model, exclusively to eligible Google AI Ultra subscribers. The update introduces real-time environment editing via natural language prompts, allowing users to modify scenes and add characters mid-exploration alongside native adaptive audio. Concurrently, Genie 3 received upgrades including extended session durations up to three minutes and 360-degree panorama grounding. The release also features a new conversational Project Genie agent and a community gallery for sharing creations, marking a shift toward fluid, responsive AI-generated environments.
Editorial responsibility
- No named human review is recorded for this page.
- Reports are grouped by semantic similarity and deterministic rules. Language models may assist titles, summaries, translation and cross-source analysis; the page reads the event directly, while its address stays stable when the title changes.
- Summary covers the current reports
Cross-source coverage
Reporting timeline
Sundar Pichai Highlights Genie 4 Interactive Prompting Feature
Google CEO Sundar Pichai shared a post on X highlighting a specific capability of the newly announced Genie 4 model. He described the feature as allowing users to type new prompts while actively exploring a world generated by the AI. This functionality enables dynamic interaction, such as introducing new characters into the environment after generation has begun. The statement serves as a direct product announcement from a key executive, detailing a user experience improvement in generative AI technology. The post emphasizes the interactive nature of the system, moving beyond static generation to real-time modification within the created digital space.
Read sourceGoogle Genie 4 Enables Real-Time Prompting and Audio Adaptation
Google has announced the availability of Project Genie 4, a new iteration of its generative AI platform. The update introduces significant interactive capabilities, allowing users to type new prompts while actively exploring a generated world. Key features include the ability to dynamically drop in characters or alter scenes mid-exploration. Additionally, the system's native audio adapts in real time to reflect these changes. This functionality is currently available for eligible Google AI Ultra subscribers. The announcement highlights a shift towards more fluid, responsive AI-generated environments where user input directly influences both visual and auditory elements instantaneously.
Read sourceGoogle Rolls Out Genie 4 for AI Ultra Subscribers with Real-Time World Editing
Google has officially launched Genie 4, the latest iteration of its Project Genie interactive world generation model, now available to eligible Google AI Ultra subscribers. This release introduces significant enhancements, including real-time environment editing via natural language prompts, native adaptive audio features, and a new community gallery for sharing and remixing creations. The update also upgrades the existing Genie 3 model with extended session durations up to three minutes, 360-degree panorama grounding, and linked world sequences. A new Project Genie agent allows users to build complex environments through conversational interactions, while Gemini models provide context-aware suggestions during exploration. While Genie 4 offers higher fidelity visuals and physics, Genie 3 remains an option for longer runs. The initiative emphasizes responsible development, incorporating SynthID watermarking and safety screenings. Google AI Pro members are invited to join a waitlist for future access.