Synthesia needs a live consent clip: what do Sume photo avatars need?
Synthesia says personal avatars need a live consent recording. Sume's photo avatar takes an HTTPS image URL, so your own release process must cover consent.
Short answer
Synthesia's avatar page says a personal avatar requires a live on-camera consent recording that cannot be uploaded or bypassed. Sume's Avatar 1.0 docs describe a photo avatar differently: you send a fetchable public HTTPS image URL, and Sume checks the URL before it submits the job. The docs do not describe a consent recording step. If you build an avatar of a real person on Sume, the consent record is yours to collect and keep.
What each side documents
The table keeps to what each page says. Synthesia and HeyGen describe consent as part of making a digital version of a person. Sume's docs describe input validation for the image URL, which is a different thing from verifying who is in the picture.
| Source | What the page says |
|---|---|
| Synthesia | Live consent recording; person in the consent video must match the avatar; 18 or older |
| HeyGen | Digital twins need verified consent from the person being recreated |
| Sume docs | image_url must be a fetchable public HTTPS image; localhost, private-network, non-HTTPS and non-image URLs are rejected |
Three kinds of Sume avatar and the consent question
Sume lets you create an avatar three ways. Only one is built from a real photograph. The others start from text or structured traits, so they have no real person behind them. That changes your consent work a great deal.
| Input | API type | Real person involved | Consent work |
|---|---|---|---|
| Prompt | prompt | No | None for likeness |
| Profile | props | No | None for likeness |
| Image | photo | If the photo is of a real person | Written release from that person |
A release you can write yourself
For a photo avatar of a real person, collect a signed release before you upload. Say who the person is, that they agree to a synthetic video version of themselves, what scripts and channels it covers, how long the permission lasts, and how to withdraw it. Store it with the avatar handle. A written release does not replace a live check the way Synthesia's does, so confirm identity yourself, for example on a call, and make sure the person in the photo is the person who signed.
- Name the avatar handle in the release.
- State the channels and the end date.
- Keep a way to retire the handle when consent ends.
- Do not use a photo of someone else without their agreement.
Why it matters now
Platform rules ask creators to disclose realistic content that makes a real person appear to say something they did not. A consent record and a script approved by the person are your evidence that the words were theirs. If you cannot show that, use a prompt or profile avatar instead and avoid the question.
Keep the paperwork proportional. For a founder who wants a presenter built from their own photo for their own company channel, a one-page release signed by that founder is enough to start. For an actor, an employee or a customer, add the limits of use and an end date. For a minor, do not use a photo avatar without a parent or guardian, and check the rules of the platform where the video will run.
Sources
Related posts
More in Sume Avatar 1.0
- Approval gate for AI avatar video: review stills before rendering
Add a sign-off step before publishing synthetic video: Sume's avatar video preview returns first-frame stills, and a later tier change needs no new preview.
- Griffin-Lite 26 of 54: error bars and a viewer test for avatars
Tavus says 48% of 54 people took Griffin-Lite for real. With 54 viewers the 95% range is 35% to 61%. A Python check and a test plan for Sume avatar clips.
- Can you buy Tavus Griffin-Lite? Presenter video options today
Tavus Griffin-Lite is a research preview for trusted testers, not a product customers can buy. What Sume ships for presenter videos in the meantime.
- Vidu Q4 audio references vs a scripted avatar: which keeps a voice?
Vidu Q4 Preview takes up to 3 audio references; Sume does not list Vidu. How a scripted avatar video keeps a presenter's face and words consistent.
Written by Sume