Security awareness training video: short clips made with AI
A security awareness training video teaches staff one security habit per clip. Make each topic a short captioned AI avatar clip and refresh only what changed.

A security awareness training video is a short clip that teaches staff one security habit, such as spotting a phishing email, protecting a password, or reporting a lost laptop. The workable build is one topic per clip with one takeaway, captions on every clip, and a yearly refresh in which you re-render only the clips whose policy changed.
With an AI avatar, each topic is one scripted render, so a refresh doesn't need a film shoot. On Sume that means one talking video per topic, one caption job per clip, and an optional join. Facts come from Generate avatar video and Video captions, read on 2026-09-28, plus Sume's current code where noted.
Which topics should security awareness videos cover?
Take the topics from your own security policy; the video makes each rule easy to remember, it doesn't write the rule. Topics that work as separate clips include:
- Phishing emails and suspicious links.
- Passwords and multi-factor sign-in.
- Reporting a suspected incident, and to whom.
- Lost or stolen laptops and phones.
- Handling customer or employee data.
- Requests by phone or chat that ask for access or money.
How do I make a security awareness video with an AI avatar?
- Create one presenter and reuse its handle for every topic; creating a reusable AI avatar shows the call.
- Write each script around one action and keep it inside the window Sume accepts: an estimated 4-60 seconds per job.
- Approve the look once. An avatar video preview returns first-frame stills before you pay for a full render.
- Render every topic with the same
avatar_handle,aspect_ratioandqualityso the series matches.qualitydefaults toplus. - Poll the job; completed results can include a public
media.sume.comvideo.
How do I add captions to each clip?
Send each finished clip to POST /v1/video-captions with its video_url, and pass the clip's script as script_text so Sume aligns the burned-in wording to your script. In current code the caption job refuses a source longer than 60 seconds or one with no audio stream; a single topic clip fits both, so caption each clip before any join. Burning captions onto a video covers the request, the styles and the alignment errors.
How do I refresh the videos each year?
Keep a list of topics with the policy version and the video URL for each; the docs tell integrations to store the Sume URL of an output. When a policy changes, edit that topic's script, render it again with the same handle, caption it, and replace its URL. The other clips stay as they are.
If you need one module rather than separate clips, Timeline 1.0 joins clips into one MP4; making an avatar video longer than 60 seconds shows how to keep each clip's speech.
Can the video track completion?
Not by itself. Sume returns the video as an MP4 file; completion tracking, quizzes, and sign-offs belong to the platform that hosts it. Converting a video to SCORM covers packaging a clip for a learning management system.
What are the limits, and what does it cost?
- In current code the avatar route speaks English only; what languages an AI avatar can speak covers other languages.
resolutionis currently720p, with one presenter per video.
| Piece | Limit | Price |
|---|---|---|
| Avatar | Created once, reused by handle | $0.95 per avatar |
| Topic clip | Estimated 4-60 seconds | $0.184/s standard, $0.245/s plus, $0.55/s max (no product image) |
| Caption job | One clip up to 60 seconds | $0.20 per job |
| Join into one module (optional) | Up to 1,800 seconds | $0.10 per output minute |
Sources
Related posts
More in Use cases
- Shopify video banner: add a looping video made with AI
A Shopify video banner is a video in a theme section on your home page. Upload the MP4 to Files, pick it in the theme editor, and make the loop with AI.
- Stream starting soon screen: make an animated loop
A stream starting soon screen is a looping video the size of your stream's canvas. How to make an animated one with AI: the loop, the words, the length.
- Text to speech time calculator: estimate, then measure
Estimate speech time as word count divided by a speaking rate, then measure the real file: Sume's TTS reports its duration with word timestamps.
- TikTok Commercial Music Library (CML): what it covers
TikTok's Commercial Music Library is a pre-cleared set of songs businesses may use free, but only on TikTok. What it covers, and music for other uses.
Written by Sume