🚨 1. Automated detection systems

Most major platforms use AI and safety filters to flag:

• Sexual messages involving minors
• Repeated adult–minor contact patterns
• Requests for sexual images
• Keywords linked to grooming behaviour
• Sharing or requesting explicit content involving minors

When flagged, content is:

• temporarily restricted, or
• sent for human review

---

👀 2. Human moderation teams

If AI flags something, trained moderators check:

• chat context (not just single messages)
• intent (manipulation, sexual framing, coercion)
• age indicators of users
• patterns of repeated contact

If grooming is suspected → escalation happens.

---

📩 3. Mandatory reporting to authorities (EU rule)

Under EU child protection frameworks (including the Digital Services Act and related child safety obligations):

Platforms must:

• preserve evidence
• report confirmed or strongly suspected child sexual exploitation to law enforcement or trusted national hotlines

In Spain, this can involve:

• Policía Nacional / Guardia Civil cybercrime units
• EU-wide coordination via Europol’s cybercrime centre (EC3)

---

🇪🇸 4. Spanish reporting channels

In Spain, reports may be routed through:

• INCIBE (Spanish cybersecurity institute)
• Policía Nacional (Grupo de Delitos Telemáticos)
• Guardia Civil (EMUME / cyber units)

Platforms cooperate by providing:

• account data
• IP logs
• message records
• device identifiers

---

🔐 5. What platforms are legally required to keep

Even if messages are deleted by users, platforms may retain:

• chat logs (for a defined legal period)
• metadata (timestamps, IP addresses, account links)
• media hashes (to detect reuploads of illegal content)

---

⚖️ 6. EU laws driving this (simple overview)

Key frameworks include:

• Digital Services Act (DSA) → requires risk mitigation for child safety
• EU Child Sexual Abuse Material (CSAM) regulations (proposed/strengthened enforcement)
• ePrivacy + national criminal codes (like Spain’s Penal Code)

These require platforms to actively prevent and report exploitation, not just react.

---

🧭 7. What usually triggers a report

A report is more likely when there is:

• sexual messaging involving a minor
• repeated adult–minor private contact
• attempts to move conversation off-platform (e.g., Telegram/WhatsApp)
• requests for images
• grooming patterns identified over time

---

🧠 Simple summary

In Spain/EU, platforms:

1. Detect suspicious behaviour (AI + moderation)
2. Review it for grooming patterns
3. Preserve evidence
4. Report confirmed or strong suspicion cases to law enforcement
5. Cooperate with police investigations (data sharing)

---