Google DeepMind Unveils Gemini Omni, New AI System for Video Creation and Editing
Google DeepMind has unveiled Gemini Omni, a new Artificial Intelligence system designed to generate and edit videos through conversational commands, marking what the company described as a major leap in world modelling and multimodal AI technology.
Speaking at Google I/O 2026, the Chief Executive Officer and co founder of DeepMind, Demis Hassabis, described Gemini Omni as a “world model” capable of building an internal understanding of reality rather than simply responding to prompts. Unlike earlier AI video generation tools trained mainly on text prompts, Gemini Omni can process combinations of images, audio, existing video and text while accurately simulating physics, motion, gravity and environmental context in generated content.
According to Google, the new system combines Gemini’s reasoning engine with DeepMind’s generative media technologies, including Veo and Genie, enabling users to edit scenes, characters and visual elements simply by describing desired changes in natural language. The company said the model also includes SynthID digital watermarking technology on all generated content to help address concerns around misinformation and AI generated media manipulation.
Google disclosed that the first version, Gemini Omni Flash, is already rolling out through the Gemini application and Google Flow for AI Plus, Pro and Ultra subscribers globally, while integration with YouTube Shorts is expected next week. Analysts say the development could significantly impact Nigeria’s fast growing creative economy, particularly Nollywood, music video production, advertising and social media content creation, by giving creators access to advanced AI powered video editing tools without the high production costs traditionally associated with professional media creation.