Google unveils Gemini Omni to push AI beyond chatbots into full-scale video creation

Google has introduced Gemini Omni, a new family of multimodal AI models aimed at transforming how users create and edit video content, marking the company’s latest push to expand artificial intelligence beyond text-based assistants and into full-scale creative production workflows.

The first model in the lineup, Gemini Omni Flash, is designed to generate cinematic videos using combinations of text, images, audio and video prompts. Unlike traditional AI video tools that largely rely on isolated prompts, Google says Omni can reason across multiple forms of input simultaneously to produce more coherent and context aware outputs.

The launch comes as competition in generative AI rapidly intensifies, with companies racing to build platforms capable of handling increasingly complex creative and enterprise tasks. AI-generated video has emerged as one of the fastest-growing segments within the broader AI ecosystems, attracting interest from creators, marketers, studios and enterprises seeking faster and more scalable production pipelines.


A key feature highlighted by Google is Omni’s conversational editing capability. Users can modify videos through natural language instructions such as changing environments, adjusting camera movements, adding visual effects or transforming artistic styles while maintaining continuity across scenes. The system also supports interactive editing allowing users to refine outputs across multiple prompts without restarting the workflow.

Google says the model demonstrates stronger “world understanding", enabling more realistic rendering of motion, lighting and environmental interactions. The company claims the system better interprets concepts such as gravity, movement and spatial consistency, areas that have traditionally remained challenging for generative video models.

Gemini Omni also builds on the momentum created by Google’s widely discussed AI image model “Nano Banana” officially known as Gemini Flash Image. The tool gained traction for its conversational image editing capabilities, allowing users to generate and modify visuals using natural language prompts while maintaining character consistency and realism. Industry observers view Gemini Omni as Google’s attempt to extend the same intuitive creative workflow from static images into full scale video generation and editing.

The platform additionally supports reference-based generation, allowing users to upload sketches, images, existing footage or audio clips that can then be transformed into stylised or photorealistic videos.

To address concerns around synthetic media and deepfakes, all the AI-generated videos created through Gemini Omni will include Google’s SynthID watermarking technology. Gemini Omni Flash will initially roll out across the Gemini app, Google Flow, YouTube Shorts and YouTube Create, with developer and enterprise API access expected later.

As AI competition increasingly moves beyond chatbots and search into creative production, Gemini Omni signals Google’s ambition to become a major player in AI-powered media creation. With conversational video editing, multimodal generation and tighter integration across YouTube and Gemini products, the company is positioning AI not just as an assistant, but as a full-scale creative engine for the next phase of digital content creation.

Nominate for ET AI Awards