CHILD PROTECTION · 7 LAYERS

Seven layers between a child and harm.

Gateway deploys seven independent detection layers to identify minors on adult platforms. The system is never punitive — it is always protective. COPPA, KOSA, and GDPR-K compliant from day one.

Part of the Candor Safety Ecosystem
CAPABILITIES
What Gateway does.
Every capability is designed to work autonomously — detecting, responding, and protecting without human intervention.
🧠
Layer 1 — Linguistic Age Analysis
Claude AI analyzes language patterns to detect writing that indicates a minor. Not keyword matching — genuine linguistic pattern recognition across vocabulary, syntax, and communication style.
📊
Layer 2 — Behavioral Patterns
Posting times, content interests, activity patterns, and session behavior. A 13-year-old's usage pattern looks measurably different from an adult's — the system learns what those patterns look like.
👤
Layer 3 — Profile Signals
Username patterns, bio content, and photo signals are evaluated. Subtle indicators that a profile belongs to a minor are detected across multiple signal dimensions.
🕸️
Layer 4 — Network Graph
Connection analysis identifies accounts linked to other flagged accounts. Minors cluster in network graphs — the system maps these patterns in real time.
🤖
Layer 5 — Proof-of-Humanity
Bot prevention ensures the account belongs to an actual person before deeper analysis. Automated account creation for exploitation is blocked at this layer.
🔐
Layer 6 — Zero-Knowledge Age Verification
Users prove they are 18+ to a trusted verifier. Their ID never touches Candor's servers. The platform stores only: age_over_18: true + a cryptographic hash of the proof.
🏳️
Layer 7 — Community Reporting
Users can report suspected minors, feeding signals back into the detection pipeline. Community intelligence supplements AI detection.
"Always protective. Never punitive. A child on the wrong platform isn't breaking a rule — they need to be kept safe."
HOW IT WORKS
The pipeline.
From detection to response — fully automated.
Account Created
All seven detection layers activate simultaneously on new account creation. No single layer is relied upon — they work together.
Monitoring Phase
If signals are detected, the account enters a monitoring phase. No immediate action — the system gathers more data across all layers.
Graduated Response
Monitoring → soft lock → suspend. Each step is protective, not punitive. The system protects a child who made it onto the wrong platform.
COPPA Compliance
Confirmed under-13 user data is automatically purged within 24 hours. Unconfirmed detection scores purge after 90 days.
COMPLIANCE & STANDARDS
Built for the real world.
COPPA
Children's Online Privacy Protection Act
Under-13 data automatically deleted within 24 hours of confirmation. Full compliance with FTC regulations.
KOSA
Kids Online Safety Act
Duty of care obligations met through active detection and graduated protective response.
GDPR-K
EU Child Data Protection
Zero-knowledge age verification ensures no unnecessary child data is collected or retained.
Ready to integrate?
Contact us for API access, enterprise licensing, or grant partnership opportunities.
[email protected]

Choose Your Plan

Child safety protection for every platform

STARTUP
$500
/month · 100K scans · 99.9% SLA
GROWTH
$1,500
/month · 1M scans · 99.99% SLA