HB Technologies LLC, operator of friendswith.ai (the “Service”), maintains identity-verification (“ID&V”) and consent procedures designed to satisfy the requirements of Epoch, the card networks (Visa, Mastercard), and the laws of the jurisdictions in which we and our users operate. This document describes how we verify the age and identity of the people who use the Service, how we obtain and record consent for any real individual who could appear in content, and how those records are retained.

1. Our model and why it matters here

Every Character on the Service is entirely computer-generated, fictional, and an adult aged 21 or older. No real performer engages in any sexually explicit conduct on the Service. As a result, our ID&V and consent obligations attach to two populations rather than to a roster of performers:

• the account holders (users) who access the Service; and
• the vendors and data sources used to train and operate our image-synthesis models.

2. Account-holder age and identity verification

• An age gate is presented before access; the user must affirm they are 18 or older (or the age of majority where higher). See our Age & Synthetic-Content Policy.
• Account creation requires a verified email address; payment, where applicable, is processed by Epoch and the card-network address-verification (AVS) and card checks serve as an additional adult-identity signal.
• We reserve the right to require step-up verification at any time — including government-issued ID and/or AI-based facial age estimation through a third-party identity provider — where activity, reports, or risk signals warrant it.
• Accounts we believe in good faith are operated by minors, or by someone who has misrepresented their identity or age, are terminated.

3. No user-uploaded content

The Service does not accept user-uploaded photographs or images. Users interact with Characters through text only; all imagery on the Service is generated by us. Because there is no user-uploaded content, no uploader consent, identity, or release records arise on the Service.

4. Consent and licensing for training data

Our image-synthesis models are trained on licensed datasets. We do not knowingly train on, or reproduce, the likeness of any real, identifiable individual, and we require our data sources to represent that their datasets are lawfully obtained and appropriately consented. Output is screened so that it does not depict real, identifiable people.

5. Records, retention, and cooperation

Verification results and moderation decisions are logged and retained for the period required by law and card-network rules. These records are made available to Epoch and to law enforcement in response to a lawful request, as described in our Law Enforcement Guidelines.

6. Contact

Questions about identity verification or consent may be sent to compliance@friendswith.ai. To report content you believe lacks consent or involves a real person without authorization, use the in-app Report tool or email trust@friendswith.ai.

7. Changes

We may update these procedures from time to time. The “Last updated” date above reflects the latest revision.