Harmful Content

Harmful content refers to material that could cause physical, psychological, emotional, or social damage to individuals or groups, including hate speech, violence, exploitation, and misinformation. In AI governance contexts, organizations must establish policies and technical safeguards to prevent their AI systems from generating, amplifying, or recommending such content, as this creates legal liability and reputational risk. This concern is central to responsible AI deployment, particularly for customer-facing applications like chatbots and content recommendation systems where oversight mechanisms are essential.

1 item

EnforcementGlobal2026-09-08

Meta's 332 CSAM Ads Expose AI Moderation as an Unreliable Child Safety Control

The Tech Transparency Project found that 332 ads containing AI-generated child sexual abuse material ran on Facebook and Instagram, many featuring real children including a European royal and a teenage influencer. Meta delayed removals for days, applied its own policies inconsistently, and reportedly continued profiting from Chinese ad resellers who ran the violating content. The findings arrive after Meta's $18 billion child safety settlement, raising serious questions about whether post-settlement controls were operationalized.

Meta CSAM content moderation child safety platform liability harmful content ad review vendor governance