Inappropriate Social Media Content

Recommendation algorithms designed to maximize engagement systematically expose minors to content that violates norms of childhood safety: sexually explicit material, eating-disorder content, self-harm content, extremist ideology, and AI-generated CSAM. The exposure is not accidental — it is the predictable output of optimization functions that treat attention as the sole metric.

Children form attachment, identity, and moral intuition through the content they consume. Algorithmic amplification of harmful content does not just expose — it shapes. The window of formation is narrow; the platforms know this and target it.

Regulatory pressure is building — the Kids Online Safety Act, state-level age verification laws — but enforcement remains fragmented. Platforms have structural incentives to resist real default protections for minors. AI-generated harmful content will further increase the scale of the problem.

Algorithmic recommendation systems optimize for engagement, which systematically pushes extreme content; child-targeted manipulation; addictive feed design; lack of guardrails; platform business model rewards harm. The youth attention market is the most valuable advertising surface; the externalities don't show on the platform balance sheet.

Platforms (ad revenue from engaged minor users), advertisers (engaged users), bad actors (reach to vulnerable populations), foreign IO actors (targeting youth radicalization).

Section 230 reform — stalled. Age gates — easily bypassed. Parental controls — minors circumvent. Content moderation — whack-a-mole at scale. Industry self-regulation — absent or performative.