Cartesia
Cartesia · Audio and voice · US · www.cartesia.ai
services/cartesia.yamlVendor data-handling terms
“Provide, operate and improve the Services, our products offerings and our business, including to generate Output and train and enhance the models that power our Services. / You may request that we not use certain categories of your Content to train our models by completing this online form .”
The policy checked contains no retention period for audio or generated content.
“Please note that the Services are designed for users in the United States only and are not intended for users located outside the United States.”
Domains and endpoints
observed means seen in Unseen deployments; vendor-documented means listed by the vendor. Vendors do not publish complete lists.
| Host | Role | Source |
|---|---|---|
| api.cartesia.ai | api | observed |
| cartesia.ai | app | observed |
Assessment
Reasoning: A voice generation and cloning API that trains on submitted content unless a form is filed, states no retention period, and declares itself for US users only. Voice recordings are biometric-adjacent data, and a cloned voice of an employee is a fraud risk in itself. An EU deployer has no basis in the policy to rely on.
Flags: Training on voice content by default · Voice cloning (impersonation risk) · Not intended for users outside the US · Assessed 2026-09-16. The assessment is Unseen's; the terms above are the vendor's.
Changelog
- 2026-09-16Documented. Training and territorial terms verified against the Cartesia privacy policy; retention period not stated.
- 2026-09-15Listed from the Unseen catalogue with observed domains.
Corrections
Pull request on GitHub, or the form below. Changes are reviewed and recorded in the changelog.