Google announced Genie 4, a real-time world model that builds on Gemini Omni and can maintain a consistent simulated environment for up to two minutes, double its Genie 3 predecessor. Users can alter scenes in real time with natural language prompts, including weather, camera movement, and new objects or characters, while the model generates ambient sound and adaptive human dialogue. Genie 4 works alongside Gemini 3.8 Flash, which breaks down prompts into steps, and is currently limited to select trusted testers in a research preview.
androidauthority.com
· 2026-10-28
Google has launched Live Avatar, a feature within Gemini 3.8 Live that pairs real-time video generation with speech models to give AI agents lip-synced, expressive on-screen personas. The avatars support 97 languages, can switch between them without quality loss, and use asynchronous tool calling to fetch data mid-conversation. The feature launches today exclusively for Gemini Enterprise subscribers, not general consumers.
androidauthority.com
· 2026-09-24
Amazon Web Services is giving select customers early access to a video-generation AI model built by The Biological Computing Company, which derives its algorithms from patterns observed in rat brain cells. The startup records how biological neural tissue processes visual information and then designs software that mimics those processes, aiming to make existing generative video models more efficient. Both companies expect a broader rollout to AWS enterprise customers to follow this limited preview.
wired.com
· 2026-09-22
GoKawiil