How are my face and voice used in AI virtual-talent ads?

On PODIANE, AI virtual talent is created only with the model’s own consent and review. Production stays within what the model has granted, the model checks and approves the result before release, and every output is labelled “AI-generated”.

Updated 2026-09-15

Consent — the model decides what and how far

AI use is split into four permissions, and the model grants them one by one.

Digital twin — a virtual talent based on your appearance

Synthetic image or video — scenes that were never shot

Voice synthesis — lines spoken in your voice

Training data — your images used to train AI

Media, term and territory apply on top. If a brand’s contract goes beyond what the model has granted, the contract is not created — the server checks it against the model’s settings.

Review — only what the model has approved goes out

The model sees every stage: the script, reference photos, voice and the finished ad, and can ask for changes.

Only results the model has approved are released. The person whose face and voice it is makes the final call.

Transparency — we say it is AI

Virtual-talent ads carry an “AI-generated” label.

Production runs on PODIANE’s own servers. Faces and voices are not sent to outside AI clouds, and model photos and profiles are closed to AI training crawlers.

What the model gains

One shoot and one recording can become ads in several languages. The model’s work reaches further without being on set every time, and the scope and term written into the contract are paid for.

The PODIANE About page shows ads made with the models’ own consent and review, in Korean, Japanese and Spanish.