Google is rolling out two major updates to Vids, its AI-powered video creation tool, that push it further into territory once reserved for professional editing suites and motion-capture studios.
The first addition, Gemini Omni, allows users to generate and edit video clips using natural language prompts and image references. Type a description, upload a rough sketch or reference photo, and Omni produces a video that matches the input. The editing layer is where it gets interesting: instead of starting over when a draft needs changes, users can describe adjustments in plain English, swap backgrounds, fix lighting, or add effects through iterative prompts. Google says this step-by-step approach means fewer abandoned drafts and faster refinement.
The second feature is more personal. Users can now create a digital avatar that looks and sounds like them by uploading a selfie and a short voice recording. Once trained, the avatar can deliver scripted messages on camera without the user ever setting up a physical shoot. Google is restricting this to users 18 and older in select regions, and each avatar is tied to the creator's Google Account to prevent impersonation.
Both features carry an invisible SynthID watermark on every generated clip, a digital fingerprint that allows platforms and viewers to verify AI-created content. This is Google's answer to the growing pressure around synthetic media transparency, and it places Vids ahead of many competitors who still treat watermarking as optional.
The pricing and availability structure follows Google's tiered model. Gemini Omni and personal avatars are available to Google Workspace Business and Enterprise subscribers. Access to avatars remains geographically limited for now.
For Google, Vids represents a bet that video, not text or static images, will become the dominant format for workplace communication and content creation. The addition of Omni editing and avatars brings it closer to that vision, though the quality gap between AI-generated and professionally produced video still exists. The question is whether that gap matters for the use cases Vids targets: quick updates, internal communications, and social content where speed beats polish.