Meta content moderation failure in Hungary violates EU Digital Services Act

This report places Meta under direct regulatory scrutiny regarding its systemic risk management obligations under the DSA. For the streaming and social video industry, it highlights the legal liability inherent in engagement-driven recommendation engines that amplify harmful content. The situation demonstrates that platform design choices are no longer viewed as neutral by EU regulators, especially when they intersect with local political hostility. As the European Commission evaluates these findings, other major platforms must prepare for stricter audits of their algorithmic safety measures in fragmented markets. Watch for the European Commission to initiate a formal investigation into Meta’s compliance with human rights protections under the DSA.

Meta faces mounting regulatory pressure across the European Union as the Digital Services Act moves from legislative framework to active enforcement. In February 2026, the European Commission opened formal proceedings against Meta under the DSA, citing concerns about addictive design features on Instagram and Facebook, marking the first time the Commission invoked its Article 66 powers against a very large online platform for systemic risk failures. That action established a precedent for the type of algorithmic accountability scrutiny that Amnesty International's Hungary findings now reinforce. The DSA requires platforms with more than 45 million monthly active users in the EU to conduct annual risk assessments covering fundamental rights, and Meta's designation as a VLOP means non-compliance can trigger fines of up to 6% of global annual revenue.

The business implications extend beyond fines into Meta's operational model for content moderation at scale. Meta announced in January 2026 that it would reduce third-party fact-checking partnerships in Europe and shift toward a community-notes model, a move that critics argued would weaken protections in markets with hostile political environments. Hungary represents a particularly acute test case because the Orbán government has systematically restricted LGBTI rights since 2020, including a 2021 law banning content deemed to promote homosexuality to minors. The Hungarian Media Authority (NMHH) has repeatedly declined to act against anti-LGBTI content on social platforms, leaving the DSA as the primary enforcement mechanism. Daniela Schnidrig, who leads Amnesty's technology and human rights research, has called on the European Commission to treat the Hungary findings as evidence of systemic failure rather than isolated moderation gaps.

Technical enforcement of the DSA is also accelerating through independent research and audit mechanisms. A March 2026 study by the AlgorithmWatch observatory found that Meta's recommendation systems amplified politically divisive content in Central European languages at rates 3.2 times higher than in English, suggesting that moderation resource allocation correlates directly with language-market size. The European Commission's Joint Research Centre published guidance in May 2026 requiring VLOPs to submit machine-readable risk mitigation plans, and Meta's initial DSA transparency report disclosed that Hungarian-language content received 68% less proactive removal than German or French equivalents. These data points provide regulators with quantifiable benchmarks for assessing whether Meta's engagement-driven ranking systems create disproportionate harm in smaller language markets, a finding that would strengthen the case for formal DSA proceedings tied to the Amnesty report.