An Examination of TikTok's Content Governance: Alignment Between Policy and Practice

Abstract

This paper explores TikTok’s platform-level content governance by assessing whether its moderation policies align with the actual outcomes of its “For You Page” (FYP) algorithm. Using a mixed-methods approach, analysing video engagement data and reviewing platform guidelines, we evaluate TikTok’s claim that content relevance, not creator popularity, drives recommendations. Results show that while lesser-known creators can gain visibility, the algorithm strongly favours entertaining content with positive emotional tones, suggesting subtle biases. A cross-platform comparison with YouTube and Instagram reveals similar trends: all three platforms amplify polished, mainstream content while often marginalizing niche or socially significant voices. Reports from marginalized creators, highlight concerns of algorithmic demotion and "shadow banning" (Eltaher et al., 2025). These findings raise questions about transparency, fairness, and accountability in algorithmic governance. With regulatory frameworks like the EU’s Digital Services Act now in play, platforms face increasing pressure to make their recommender systems auditable and equitable. The paper concludes by calling for clearer policy enforcement, independent algorithm audits, and stronger safeguards to protect content diversity and creator equity.