Google has decided that appearing in your own corporate videos is now far too much effort and has upgraded Google Vids so an AI-generated version of you can do the talking instead. The company is rolling out “personal avatars” that create a digital representation of a user from a selfie and a short voice recording. Once the avatar has been generated, users simply type a script and their synthetic stand-in delivers it without the need for a camera, microphone, lighting setup or any of the other inconveniences traditionally associated with recording a human being.
Google Vids has already been offering AI presenters, but these were pre-made characters rather than digital copies of the person using the service. The latest system brings the technology rather closer to home, allowing managers, trainers and executives to produce video updates featuring something that looks and sounds like them while they are presumably busy attending another meeting about productivity.
Alongside the new avatars, Google is bringing its Gemini Omni model into it’s Vids platform. Users can start with a natural-language description, add reference images such as photographs or rough sketches, and have Omni combine the material into a generated video clip. The model can also work with existing footage, including video recorded on a phone, giving users a way to build or modify content without starting from scratch.
Editing is handled through conversational prompts rather than conventional video-production controls. Users can ask Google Vids to replace a background, adjust lighting, add effects or make other changes simply by describing the desired result. Google Omni also supports iterative editing, allowing a clip to be refined one step at a time rather than regenerated in full whenever the AI puts the office furniture on the ceiling or gives someone an unexpected collection of additional fingers.
The additions push Google Vids beyond its original position as an AI-assisted workplace presentation service. It increasingly resembles a broader video-production platform aimed at creating training material, internal announcements, sales presentations, customer messages and the sort of executive updates that could previously have been an email.
That also puts Google into more direct competition with dedicated AI-avatar and video-generation services. Right now the main advantage Google has, is that the Vids platform already sits inside their Workspace ecosystem, where companies store their documents, presentations and other source material. Instead of persuading businesses to adopt another platform, Google can simply bolt the technology onto software they are already paying for.
There are some restrictions currently, at least on paper, in that the personal avatars are tied to the account holder’s Google Account and may only use that person’s likeness. Access is currently limited to adults in supported regions, although the company has not published a complete list of where the feature will be available. Every generated clip that the system generates, will also contain Google’s invisible SynthID watermark, intended to identify the material as AI-generated. That should provide some traceability, although an invisible watermark is unlikely to stop viewers from believing that a convincing synthetic executive really did approve the latest restructuring plan.
The practical appeal is obvious. Businesses can generate consistent training videos, turn routine announcements into repeatable presentations and update scripts without dragging the original speaker back in front of a camera. It also means a person’s digital likeness could continue delivering polished company messages long after the real employee has stopped caring which quarter the synergy roadmap is targeting. Gemini Omni and personal avatars are rolling out in Google Vids to Google AI Pro and Ultra subscribers, along with eligible Google Workspace business customers. There is no indication that free users will receive the full feature set.
