Inspired by this story? Turn the idea into your own AI art in seconds — free to start, no card required.
Google Vids is getting two major upgrades: personal AI avatars that let Workspace users star in videos without stepping in front of a camera, and Gemini Omni-powered tools for generating and editing footage directly from text prompts and reference images.
According to Google's official blog post, users record a short consent video to train their avatar, after which the system can generate clips of their digital likeness speaking scripted content. The practical upshot: a creator or marketer can produce a talking-head explainer, product walkthrough, or social clip without scheduling a shoot, managing lighting, or doing multiple takes. For AI-art creators who regularly produce tutorials or showcase reels, that's a meaningful reduction in production friction.
The avatar is tied to a specific Workspace account and requires explicit opt-in, which addresses at least some of the consent concerns that have plagued similar features elsewhere — most recently Meta's now-pulled Instagram AI Muse, which generated images of public accounts without permission.
The Gemini Omni integration is the more broadly useful capability for creators who aren't interested in avatar-based content. Users can describe a scene in a text prompt, optionally supply a reference image to anchor the visual style, and have Vids generate footage — all inside the same Workspace environment where they already handle Docs and Slides. Editing works the same way: describe the change, and Omni applies it.
That reference-image input is worth paying attention to. It means you can feed in a character design, a brand illustration, or a mood-board frame and use it to constrain the output — closer to how experienced prompt engineers already work with image generators on platforms like Charmloop's generator. The gap between static AI image creation and AI video creation keeps narrowing, and tools like this accelerate that convergence.
For context on where AI video generation is heading more broadly, the world models explainer at Charmloop covers why the underlying architecture matters for anyone building generative video workflows.
Google Vids has always lived inside the productivity suite rather than the creative one, which means its primary audience is business users, not dedicated AI-art creators. But the Gemini Omni addition starts to blur that line. If you already use Workspace for client communication or project management, having a capable text-to-video tool in the same environment — with no separate subscription or API key — lowers the barrier to producing video content considerably.
The avatar feature is more niche but potentially high-value for creators who produce regular educational or branded content and want a consistent on-screen presence without the overhead of filming. It also sidesteps the camera-shyness problem that keeps many technically skilled creators off video platforms entirely.
As TechCrunch reports, both features are rolling out now to Workspace users. Google hasn't specified which Workspace tiers get access at launch, so creators on lower-tier plans should verify availability before building it into a workflow. If you're already experimenting with AI video tools, the Charmloop guides section covers prompting techniques that transfer directly to reference-image-based generation — including the kind of style-anchoring that Gemini Omni now supports.
