Sextortion — Instagram
California · 4/5/2026
Grooming — Discord
Texas · 4/4/2026
Self-harm content — TikTok
Florida · 4/3/2026
Off-platform migration — Roblox → Telegram
New York · 4/2/2026
Sextortion — Snapchat
Illinois · 4/1/2026
TeenAegis cross-validates threat intelligence across six frontier AI models to eliminate single-model bias, hallucination risk, and coverage gaps. Each model contributes distinct signal types. Consensus scoring requires agreement from at least 4 of 6 models before a threat pattern is elevated to the TAIM stack.
How kids are led into harm — Algorithmic Drift · Content Adjacency · Platform Penetration
Ofcom 2024 §4.2 · Stanford IO 2023
Stanford IO 2023 · DSA Transparency 2024
Pew Research 2024 · Common Sense 2023
What this proves: Harm is not random. It is engineered proximity. Platforms design recommendation systems that accelerate the journey from safe to harmful content — average drift time is 11 minutes on the worst-performing platform.
Entrapment velocity accelerating — Bay Area cohort, Discord → Telegram migration detected
NCMEC pattern analysis
Platform response lag exceeding 72-hour threshold on X — repeat offender survival rate 83%
IWF 2024 · FTC records
LGBTQIA+ targeting efficiency spike — Instagram identity signal amplification +4.7 pts
Thorn 2023 §3.1
Initiate open conversation about platform migration patterns — average awareness lag is 4.7 days
Activate sextortion response playbook for Bay Area cohort — incident clustering detected
Escalate X repeat offender survival rate to FTC — 83% survival rate exceeds negligence threshold
Escalation to off-platform coercion within 30 days
Escalation probability reduced by 56%
Escalation probability reduced by 74%
Incident Timeline
Sextortion — Instagram
California · FBI IC3 pattern analysis
Grooming — Discord
Texas · NCMEC CyberTipline
Self-harm content — TikTok
Florida · Ofcom 2024 §4.2
Off-platform migration — Roblox → Telegram
New York · Europol IOCTA 2024
Sextortion — Snapchat
Illinois · Thorn 2023 §3.1
Evidence Archive — Litigation-Ready Sources
TeenAegis Intelligence Model (TAIM) v1 — Methodology Statement. All indices are computed from publicly available, government-published, peer-reviewed, and court-verified datasets. Source citations are available to verified researchers, regulators, and legal counsel on request. Scoring methodology is proprietary to TeenAegis and is not disclosed in this interface. All data is presented for intelligence and policy purposes. Nothing in this interface constitutes legal advice. © TeenAegis 2026. Unauthorized reproduction, scraping, or systematic extraction of this data is prohibited.
Entrapment via identity-affirming grooming
Algorithmic drift into self-harm content
Off-platform migration to encrypted channels