All AI news
AI NewsGemini

Google Vids can now turn your photo into an AI avatar

Google Vids now lets you upload a photo and turn it into an AI avatar that reads your script. Each avatar video can run up to 60 seconds. YouTube requires a label when AI makes a real person appear to say something they didn't.

Jessica Wells··5 min read

Your next video spokesperson might be a photo of you. No second take, and apparently no argument about the lighting.

What Google shipped

Google's Workspace Updates post says you can now upload images to create custom avatars in Google Vids, on top of the existing option to build one from scratch. You choose an image from your computer, Drive or Photos, refine it inside Vids to clean up the background or lighting, and save it to the project so you can reuse it consistently in a video.

The same post gives the rollout. For Rapid Release domains, a gradual rollout began October 30, and the feature can take up to 15 days to show up. Scheduled Release domains get it starting November 20. There is no admin control for the feature. It's listed for Business Starter, Standard and Plus, the Enterprise plans, and personal Google accounts with Google AI Pro or Ultra, among other editions.

How it works, per Google's help page

The Vids help page walks through it:

  • The photo: Google recommends a high-quality 16:9 landscape image with a clear, well-lit human face and a neutral background.
  • The voice and script: you pick a voice from a list, type the script, and the avatar speaks it in your scene.
  • Reuse: the avatar is saved in that Vids project. To use it in another file, you make a copy of the project.
  • The spokesperson option: if you build an avatar from scratch instead, you can add your business logo so the avatar acts as a spokesperson for your business.

The limits matter for planning. The help page says many users can generate up to 25 avatars a month, some plans share that limit with AI video clips, and each avatar video can run up to 60 seconds. Google suggests training videos and presentations. Fair enough. The same avatar could also end up in your next ad.

Marketing insight: Use avatars first where the job is explaining, like onboarding, how-to clips and answers to common questions. Save your real face for the videos where trust decides the sale.

What Google didn't say

Neither the post nor the help page gives quality data, usage numbers or examples of results. Neither says whether a Vids avatar carries a watermark. And neither spells out whose photo you're allowed to upload. The help page does point to where the answer lives: it says the avatar feature is subject to Google's Generative AI Prohibited Use Policy.

The rules that apply anyway

That policy applies to Google products that refer to it, Vids avatars included. Among other things, it says not to:

  • Impersonate an individual, living or dead, without explicit disclosure, in order to deceive.
  • Use personal data or biometrics without legally required consent.
  • Facilitate misleading claims of expertise in sensitive areas such as health, finance, government services or the law, in order to deceive.
  • Claim generated content was made solely by a human, in order to deceive.

If the video goes on YouTube, there's another rule. YouTube's disclosure policy requires creators to disclose realistic AI-generated or altered content, including content that makes a real person appear to say or do something they didn't. You do it with the "AI use" setting in YouTube Studio, and YouTube adds a label for viewers.

So an avatar of your own face, reading your own words, with your own sign-off, is one thing. An avatar of your office manager, made from a team photo, saying things they never approved, is a very different thing, and a very awkward staff meeting. If you work in healthcare, finance or law, read the expertise line in that policy twice.

Spotting AI video is getting easier

In its October recap, Google said it is expanding SynthID Detector globally in English, so anyone can check whether an image, video or audio clip was generated with AI. Google says the detector also identifies AI content from partners including OpenAI, NVIDIA, Kakao and Apple, and that its built-in verification tools in Search, Chrome and the Gemini app handle over a million requests a day.

Google doesn't say whether the detector catches Vids avatars. But viewers are getting more ways to check what was made with AI. Telling them first is a much easier conversation.

What to do this week

  • Pick the right first video. Try an internal training clip or a short how-to before anything a customer sees.
  • Get a yes in writing. Only upload a photo of someone who has agreed to have an avatar made from it.
  • Label it. Use YouTube's AI setting where it applies, and say in the caption that the presenter is an AI avatar.
  • Review the script like an ad. The avatar will say anything you type, including claims your business can't back up.
  • Plan around the limits. Up to 60 seconds per avatar video means short, focused scripts.

A face isn't the same as a person

Avatars make video cheaper to produce, and your competitors get the same tool. That makes the real thing stand out more: a founder on camera, a real customer, a real answer.

Thinking about AI video for your business? Talk to our founders below. See you in the next edition.

Sources

Sources, checked November 2, 2026. Every claim above links to its original source. We summarize and comment; the reporting belongs to the outlets named here.

AI consulting for businesses

Want AI working for your business, not just in the headlines?

Our founders look at how you sell today, then show you where AI earns its keep, where it is still a toy, and what to do first. For businesses of any size.

Let's talk