Transparency at Secret Desires AI
Last Updated: March 9, 2026
Trust requires visibility. This page outlines how we moderate content, respond to legal requests, and invest in platform safety — including independent verification by third-party security firms. Our first quantitative transparency report publishes Q2 2026.
Content moderation#
How we moderate#
Our moderation system operates in multiple layers. Automated detection combines keyword filtering with AI-powered content classification to flag potential violations in real time. Every flagged item is scored across multiple violation categories using confidence thresholds — not simple keyword matches alone.
Complex or borderline cases are escalated to trained human reviewers. Third-party compliance verification provides an additional layer independent of our internal systems.
What we action against#
CSAM
Child sexual abuse material. Zero tolerance. Detected and reported.
Celebrity content
Unauthorized use of real individuals' likenesses.
Deepfakes
Non-consensual synthetic media depicting real people.
Bestiality
Sexual content involving animals.
Non-consent
Content depicting non-consensual scenarios.
Incest
Content depicting incestuous scenarios.
Legal requests#
How we handle legal requests#
When we receive requests from government or law enforcement agencies, we evaluate each for legal validity before responding. We comply with lawful requests that meet jurisdictional requirements. We push back on requests that are overly broad or legally insufficient. We notify affected users when legally permitted to do so.
Safety and compliance investments#
SOC 2 Type I & II
Independent compliance audit by Delve. Initiated February 2026. Covers security, availability, and confidentiality controls.
Penetration testing
Independent vulnerability testing by Casco Security and BSK Security. External teams actively probe our systems.
Delve Trust Hub
Public, third-party verified compliance profile — coming soon.
Moderation investment
Continuous improvements to automated detection accuracy, ongoing moderator training, and capacity expansion.
Reporting timeline#
Our first quantitative transparency report — covering moderation actions, legal requests, and safety metrics — publishes Q2 2026. Subsequent reports will follow quarterly, each covering the previous three-month period.
Methodology#
Transparency reports cover rolling three-month periods. Data is sourced from our internal moderation and legal compliance systems. "Actioned" accounts include warnings, temporary suspensions, and permanent bans. Content removals are categorized by violation type. Legal requests are reported by type and jurisdiction.
For questions about our transparency practices, contact [email protected].
Security & Safety
AES-256 encryption, Cloudflare protection, and the full technical details.
See full security details →
