Google adds personal AI avatars to Vids for Workspace users

Google is giving its Vids app a more personal AI feature: users can create videos featuring a digital version of themselves. For investors watching Alphabet, the update shows Google continuing to fold generative AI into Workspace, its paid productivity suite, where AI tools can support both retention and upselling.

Google said Thursday that Vids will let eligible users create a custom avatar that resembles them and uses their voice. The avatar is generated from a selfie and a voice recording uploaded by the user, according to the company.

Vids began as an AI-assisted tool for workplace presentations, but the new features move it closer to a broader video production product. In plain terms, Google is trying to reduce the amount of filming, editing, and production work needed to make a polished internal video, such as a company update or training clip.

How the new avatar feature works

Google said the personal avatar is linked to the account holder’s likeness and connected to that person’s Google account. The company also said the content is invisibly watermarked with SynthID, Google’s technology for marking AI-generated media in a way that is not visible to viewers.

Access will not be open to every user. Google said personal avatars are limited to people in certain regions and to users who are at least 18 years old.

The setup matters because AI likeness tools can raise identity and consent concerns. By tying avatars to the account holder and adding an invisible watermark, Google is presenting the feature as a controlled workplace tool rather than an open-ended celebrity or public-figure imitation product.

Gemini Omni comes to Vids

Google also said it is bringing Gemini Omni to Vids. Gemini Omni is Google’s multimodal AI model, meaning it can work across different kinds of inputs, including text, images, audio, and video.

In Vids, users will be able to generate videos from a written prompt and uploaded reference images, according to Google. The model combines those inputs to produce a video based on the user’s request.

Google said Gemini Omni can also help edit existing phone-shot videos. The company listed examples including changing a background, correcting lighting, and adding effects.

The company also said Omni now supports step-by-step editing. That means users can make additional changes during the creation process instead of restarting the video from the beginning each time they want to revise something.

Why it matters for Google’s AI strategy

The update puts Vids in a more crowded part of the AI software market. AI avatar and video-generation tools from companies such as HeyGen, Synthesia, Captions, and D-ID already offer ways to create presenter-style videos without a traditional shoot.

Google’s advantage is distribution. Vids sits inside Google Workspace, which already serves businesses using tools such as Gmail, Docs, Drive, and Meet. Adding AI video creation there gives Google another way to make Workspace feel more complete for teams that already pay for productivity software.

The launch also comes after OpenAI’s Sora app shut down, according to TechCrunch. Google’s move suggests the company still sees demand for AI video tools, especially when packaged for business use rather than purely consumer experimentation.

Google did not disclose pricing, a launch timetable beyond the announced update, or usage limits in the announcement cited by TechCrunch.

This story draws on original reporting from TechCrunch.