Google Vids Personal Avatars: AI Video Where You’re the Star — No Camera Needed

AI NOTES · EN ENGLISH EDITION

Google Vids Personal Avatars: AI Video Where You're the Star — No Camera Needed

KO · 한국어 / EN · English BILINGUAL PAIR

On July 16, 2026, Google announced two major updates to Google Vids: Google AI Omni, which lets you generate and edit videos with text prompts and image references, and personal avatars, which let you create a digital version of yourself from a single selfie and a short voice recording.

What changed with this update is that users can create a digital avatar that looks and sounds like them to deliver a message without setting up a camera.

Two updates, one direction

Google AI Omni: generate and edit with text prompts

Google Vids already offered text-to-video generation through Veo 3.1, rolled out to all users in February. Google AI Omni takes this further.

• Combine text and images as input. Describe a scene — “a person walking on a beach at sunset” — and add a reference photo or rough sketch. Omni blends both into a generated clip.

• Edit with text prompts. Type “swap the background to a forest,” “warm up the lighting,” or “add effects,” and Omni applies the change.

• Step-by-step editing. You can refine specific parts of a video without starting over from scratch.

Text and image references become a generated clip that can be refined through step-by-step text-prompt edits.

Personal avatars: one selfie, and you’re ready to film

The more striking update is personal avatars. Upload a selfie and a short voice recording, and Google Vids creates a digital avatar that looks and sounds like you. From there, type what you want to say — your avatar delivers the message. No camera setup, no re-recording.

Google has placed clear boundaries around this feature:

• Account-holder likeness only. You cannot create an avatar of someone else.

• Limited to certain regions and users 18 or older.

• Available to Google AI Pro, Ultra subscribers, and Google Workspace business customers.

A separate update: custom AI avatars can perform directed actions

The June 17 Workspace update added text-prompt action direction to custom AI avatars. This was a separate feature update from the personal avatars announced on July 16.

Type instructions like “walk, talk, and use objects,” and the custom AI avatar follows them in the generated clip. Upload a brand logo or a location image, and the custom avatar moves within that space.

As described in the June update, AI avatars and voiceovers support 24 languages, including Korean, while custom avatars offer 30+ Google AI Audio voices. These details should not be assumed to apply identically to the personal-avatar feature announced in July.

The custom AI avatars announced in June support voice selection and text-directed actions, while generated clips carry SynthID.

Transparency: SynthID watermarking

Every AI-generated clip includes an invisible SynthID digital watermark. It isn’t a visible label, but it can help verify that a clip was created with AI. Google describes it as a content-transparency measure.

Who can use it

Google AI Pro

Google AI Omni ✅

Personal Avatars ✅ (select regions, 18+)

Google AI Ultra

Google AI Omni ✅

Personal Avatars ✅ (select regions, 18+)

Workspace Business

Google AI Omni ✅

Personal Avatars ✅ (select regions, 18+)

Free Google Account

Google AI Omni ❌

Personal Avatars ❌

Actual availability can vary by country, language, and account. Monthly generation limits also vary by plan. For many users, this means up to 25 AI avatar video generations per month, with each avatar video capped at 60 seconds.

Why this matters

For users who found AI video tools difficult, prompt writing and repeated revisions could be a barrier.

This Google Vids update lowers the barrier in three concrete ways:

• Camera setup becomes optional. You can appear in videos without recording yourself on camera.

• Editing shifts toward text prompts. Type “brighten this,” “change the background,” or “add effects” instead of rebuilding the clip.

• Repeat production gets easier. Create your avatar once, then change the script to produce multiple videos.

This is practical for educators, brand marketers, customer communications, and individual creators alike.

What to watch out for

• Avatars are not real people. They are AI-generated representations.

• AI responses can include mistakes. Google explicitly notes this.

• Personal avatars are not yet available worldwide. Regional restrictions apply.

• Free accounts are excluded. A paid Google AI or Workspace plan is required.

• The watermark is invisible. SynthID is a non-visible digital watermark — viewers cannot see it with the naked eye.

• A visible disclosure may still help. Because SynthID is invisible, public or sensitive videos may benefit from a separate on-screen or written AI-generation label.

• Selfies and voice recordings are personal data. Check your personal account or organization’s privacy and content policies before uploading them.

Bottom line

Your face, voice, and script can become video without a camera setup or repeated recording.

This Google Vids update marks a shift from AI video as a technical demo to AI video as an everyday tool. Personal avatars, in particular, offer a practical path for anyone who wants to create video content but doesn’t want to be on camera.

Regional limits and subscription requirements remain, but the direction is clear: the center of gravity in video production is moving from equipment and editing skill toward ideas and scripts.

References

• Create, edit and star in videos with two Google Vids updates — Google Blog, 2026-07-16

• Use AI avatars in Google Vids — Google Docs Editors Help

• Enhanced AI avatar features and capabilities in Google Vids — Google Workspace Updates, 2026-06-17