Skip to content

Data Protection

Last updated 24 August 2026

How we handle the data you entrust to us, in practical terms — written for the person doing a vendor review.

Draft — not yet legally reviewed. The {{PLACEHOLDER}} values (legal entity, jurisdiction, contact email) still need filling in, and a solicitor should review this before you rely on it.

Data flow for an image render

1. You upload a source image over TLS. It is written to private storage, keyed to your user id.

2. One credit is debited inside a database transaction before generation starts, so concurrent requests cannot overdraw the balance.

3. The image and the composed prompt are sent to the configured AI provider.

4. The result is written to private storage and linked to your account, with an expiry timestamp 7 days out.

5. If generation fails, the credit is refunded automatically and the failure is recorded against the render.

Access control

Every stored file is namespaced by user id, and the file-serving route verifies ownership before returning bytes.

Requests for a file you do not own return 404 rather than 403, so the existence of another user's file is not disclosed.

Path traversal attempts are rejected before touching the filesystem.

Sub-processors

AI image generation provider (configurable — the operator selects one).

Anthropic, for AI-written ad copy.

Hosting and object storage provider.

A current list is available on request.

What we do not do

We do not train models on your uploads.

We do not make your content public by default. Publishing is an explicit, reversible action you take per render.

We do not embed third-party advertising or analytics trackers.

Breach handling

If we become aware of a personal data breach that is likely to result in a risk to your rights, we will notify affected users and the relevant supervisory authority without undue delay. Report a suspected issue to {{CONTACT_EMAIL}}.

Questions about this document? Contact us.