Google Unveils New Gemini Omni AI System for Video Creation

Google has introduced Gemini Omni — a new AI system for creating and editing video through a simple text or voice dialogue.

The announcement was made at Google I/O 2026. The first version, Gemini Omni Flash, is focused on video editing and allows users to work with video without complex editors — simply describe what needs to be changed or created.

The system supports multiple data types: text, images, audio, and video. Users can ask it to change the background, lighting, or object movement, and the model handles it automatically while preserving scene coherence and realism.

One of the key features is "conversational editing," where changes are made through a dialogue with the AI. The system also highlights improved physical accuracy in video generation — object movements and interactions look more natural.

The technology is already being integrated into Gemini, YouTube Shorts, and other Google services. All AI-generated content will be labeled via SynthID to make it clear that it was produced by artificial intelligence.