TL;DR
- Google Vids has received a major AI overhaul with Gemini Omni, making it easier to create and edit videos and explain what you want in plain language.
- You can create your personal AI avatar by uploading a single selfie and a short audio recording.
- To prevent abuse, personal avatars are limited to the likeness of the account owner.
Editing videos usually involves a long process of fixing lighting, changing backgrounds, and taking multiple retakes to get the perfect shot. Now, Google is rewriting its strategy with a major update to Google Vids that puts its AI work to work.
The biggest new feature to Vids is Gemini Omni, Google’s multimodal video generation model that debuted at I/O 2026. According to Google’s blog post, Vids lets you create or edit videos from natural language prompts, whether you want to start from scratch with text and images or edit existing clips. Just skip the timeline and tell Gemini to change the background, improve lighting, and add visual effects with conversation prompts.
Google also lets you add a personal avatar to your Vids, allowing you to appear in videos without having to be in front of the camera. This feature was previously available to all US Google account holders. Upload a selfie and record a short audio sample, and Vids creates a digital version of you that can deliver any script you enter. This feature is only available with Google Accounts and Images. Additionally, as a safety precaution, Google restricts this feature to users 18 and older in supported regions. Additionally, all AI-created clips have an invisible watermark called SynthID that can be used to verify their source.
Google is also giving users more control over AI narration. You can now add commands enclosed in parentheses, such as “”.[excitedly]According to the company, you can change your avatar’s emotions, pace, startup sound, and more by changing prompts such as “” directly in your script. When you type in an opening bracket, Google Vids will suggest available steering tags, and if you don’t want to add them manually, a new “Apply Audio Tags” option will automatically insert the appropriate cues throughout your script.
The Voice Steering tool is now available in both immediate and planned release domains and supports Google Workspace Business and Enterprise editions, Education Plus, nonprofits, individual users, and personal Google Accounts (including Google AI Pro and Ultra subscribers). This feature is available automatically, so administrators don’t need to enable anything.
Thank you for joining our community. Please read our comment policy before posting.
