How to Make Videos with Gemini in Google Vids

To make videos with Gemini in Google Vids, open vids.google.com, click Create and choose Help me create to start with a Gemini prompt. Describe the topic, audience and tone. Gemini drafts a script, storyboard and scenes with stock media, AI voiceover and, on eligible plans, an AI avatar presenter. You then edit the scenes, swap media, refine the voice and share the video from Drive.

This guide walks through the workflow end to end, then covers where AI-generated video actually earns attention and where it falls flat. If you build content for a brand, an agency or a client, the second half matters more than the first.

What is Google Vids and where does Gemini fit in

Google Vids is a video creation app inside Google Workspace, sitting alongside Docs, Slides, Sheets and Meet. It uses the same collaborative editing model. Multiple people can work on the same video at once, comments live inside the file, and the finished video lives in Drive.

Gemini is the AI layer built into Vids. It handles four jobs: drafting a full storyboard from a prompt, generating images and B-roll clips on demand, writing and rewriting on-screen copy, and producing voiceover in a range of AI voices. On eligible plans, it can also place an AI avatar presenter on screen, reading from a script you approve.

For anyone building content programs, the shift is bigger than it looks. Video used to sit outside the daily doc-and-slide flow because production had its own tooling, its own turnaround and its own cost line. Vids collapses that. A product manager, a solutions engineer or a marketer can now brief and ship a rough video in the same window they use to write a memo.

What you need before you start

• Google Workspace plan with Vids access (Business Standard and above, plus Google AI plans). AI avatars and full generative features sit on higher tiers.

• A clear brief: topic, audience, one thing you want the viewer to remember, and where the video will be watched.

• Brand assets: logo, brand colors, one or two brand fonts, and any pre-approved imagery.

• A script or bullet outline if the topic is technical, regulated or legally sensitive. Do not let Gemini invent claims in those spaces.

How to make videos with Gemini in Google Vids step by step

Step 1: Open Google Vids and start a new project

Go to vids.google.com or open the app launcher inside any Workspace product and pick Vids. Click Create in the top left. You get three options: start from a blank video, start from a template, or use Help me create to trigger the Gemini flow.

Step 2: Prompt Gemini with a clear brief

Pick Help me create. A prompt box opens. This is the highest-leverage moment in the whole workflow. A vague prompt produces a generic storyboard. A specific prompt produces a draft you can actually use.

A useful prompt names four things: what the video is about, who it is for, what tone it should carry, and what action the viewer should take at the end. For example:

"Create a two minute internal explainer for our sales team on how our new billing feature works. Tone is confident and plain-spoken. End with a call to review the enablement doc.", 

Gemini returns a storyboard with scene-by-scene visuals, on-screen text, a suggested voiceover script and an estimated length. You can regenerate, edit the prompt or accept and open the project.

Step 3: Review the storyboard scene by scene

Vids opens with the video timeline at the bottom, a scene panel on the left and the preview in the center. Walk the storyboard scene by scene. For each one, check three things:

1. Is the on-screen claim accurate. Gemini will confidently invent statistics, product features and integration names that do not exist. Rewrite anything you cannot verify.

2. Is the visual on brand. The default stock and generated images are generic. Replace hero visuals with your product screens, real footage or brand imagery.

3. Does the scene earn its runtime. Cut anything that repeats what the previous scene already said.

Step 4: Swap in your media

Add your own footage from Drive, upload from device, or use Generate to create fresh images and short video clips from a prompt. For product-led scenes, screen recordings almost always outperform generated visuals. For lifestyle or ambient B-roll, generated media can hold its own. Real UGC still outperforms both for social ad creative, which is why we run UGC production at scale as a separate service line for brands that need it.

Step 5: Refine the voiceover or record your own

Every scene has a script and a voice. You have three choices.

• AI voiceover. Pick a voice, language and speed. Good enough for internal, educational and top-of-funnel explainer content.

• AI avatar. On eligible plans, place a photorealistic presenter on screen reading your script. Useful for founder-style updates and onboarding, weaker for anything that needs emotion.

• Your own voice. Record directly in Vids or upload audio. This is the right call for anything customer-facing where trust matters.

If you plan to build a library of avatar-led explainers, that program design sits closer to our AI avatar and AI video production work than to a one-off Vids project.

Step 6: Style the video with your brand

Open the theme panel and set brand colors, font and logo placement. Vids will not do full brand design for you, so if the brand system is thin or inconsistent, fix that first through proper branding and design work. A weak brand system produces weak-looking video no matter which tool renders it.

Step 7: Add transitions, timing and captions

Trim scenes on the timeline, snap transitions where the pacing needs a beat, and turn on auto-generated captions. Around 85% of social video is watched with the sound off in early scroll, so captions are not optional if the video will live outside Drive.

Step 8: Share, export or embed

Click Share to send a Drive link with view, comment or edit permissions, exactly like a Doc. Download as MP4 for upload to YouTube, LinkedIn, an LMS or an ad platform. For a website, embed the Drive link or host the MP4 on your CDN.