How QLumis protects what you share.

Encryption

Memory content is encrypted at rest using QLumis-managed encryption secrets. Sensitive vendor credentials are encrypted separately.

Transport security

All traffic between the browser and QLumis is served over HTTPS with strict transport security and a strong content security policy.

Authentication

Accounts are protected by password authentication with optional OAuth providers and a bot-deterrence check at sign-up.

Consent and identity controls

Voice and likeness profiles beyond your own are gated by consent or appropriate authority, enforced inside the app. Consent can be reviewed or revoked, and a companion always identifies itself as QLumis. Voice and likeness are treated as identity data, not ordinary settings.

Vendor routing and no-training posture

QLumis is designed to route sensitive companion content through approved providers and configurations that do not use your content for vendor-side model training. Workflows that cannot support this posture are not used for sensitive memory features.

Auditability

Where enabled, QLumis maintains audit logs of significant account actions — including biometric enrollment and consent changes — so sensitive events leave a record.

Data deletion

You can delete memory, voice, and likeness data from inside the app, subject to the platform’s privacy and retention policies. The design goal is to avoid unnecessary data retention.

Operational logging restraint

Operational logs are written to debug and protect the service, not to capture conversation content. The posture is to log what is needed to keep QLumis running safely and no more.

Responsible disclosure

If you believe you’ve found a security issue, please email [email protected] with details. We appreciate disclosures that give us time to investigate before publication.

What this page is not

This page is not a formal compliance certification. Specific regulatory claims for a given workflow should be confirmed with our team in writing.