Can the DSA adequately address online hate speech targeting young people, or does the EU need targeted hate-speech legislation?

► President
I open this on a question from the Council of Europe's No Hate Speech Week: is online hate speech Europe's biggest threat to young people's involvement in democracy? The framing invites us to think about causation and remedy. If hate speech is indeed a primary barrier, what tools does the EU have, and are they adequate? The institutional question is sharp: the Digital Services Act already addresses illegal hate speech and systemic risks to minors and democracy through its systemic-risk framework. But illegal hate speech under national laws varies widely across the 27 — what is prosecutable hate speech in France is protected speech in Germany. The DSA enforces against illegal content as defined by member-state law, but does not create an EU-wide hate-speech crime. The question for the College is whether the DSA's existing toolkit is sufficient, or whether we need targeted EU hate-speech legislation that harmonises the definition and creates EU-level criminal liability. The legal base would be Article 83(1) TFEU, requiring unanimity in the Council.
[EUR-Lex — verify DSA Regulation (EU) 2022/2065 Arts. 34-35; TFEU Art. 83(1)]

► EVP · Democracy
This portfolio leads jointly with Justice. Online hate speech targeting minors, religious minorities, LGBTQ+ individuals and political opponents is widespread, algorithmically amplified, and a documented barrier to youth democratic participation. Young people report self-censoring, withdrawing from platforms and losing trust in democratic institutions. The harm is real. What the DSA does: Articles 34 and 35 require VLOPs to assess and mitigate systemic risks including illegal hate speech as defined by member-state law. The Commission can enforce. The DSA does not require removal of lawful-but-hateful speech; it requires transparency, risk assessment and mitigation within national illegality. The problem: national definitions are fragmented — France criminalises hate speech on religion, race, ethnicity, disability, sexual orientation and gender identity; Germany has the NetzDG Act; the UK has the Online Safety Bill. But Hungary's law on content promoting homosexuality suppresses LGBTQ+ speech; Poland's defamation law chills political criticism. A platform removing illegal hate speech in Hungary may be removing protected speech in France. Recommendation: strengthen DSA implementation via robust audits and an EU-wide definition of high-risk hate speech as a systemic-risk trigger — not criminalising it, but guiding mitigation. On targeted hate-speech legislation, the legal base is Article 83(1) TFEU, requiring unanimity. Given member-state divergence, unanimity is politically impossible. The better path: DSA enforcement plus national prosecution.
[EUR-Lex — verify DSA Arts. 34-35; TFEU Art. 83(1); national hate-speech laws]

► Commissioner · Justice
Support for DSA enforcement and national prosecution. The Charter protects both freedom of expression under Article 11 and the rights of the child under Article 24. Hate speech targeting minors can violate the child's dignity and right to develop free of intimidation — that is enforceable. On a targeted EU hate-speech law: the legal base is Article 83(1) TFEU, requiring unanimity. Given member-state diversity on what counts as hate speech, the unanimity requirement will stall any proposal. DSA plus national prosecution is the workable path. Within that, I want clear guidance on proportionality: removal of hate speech must be proportionate to the harm, and the Charter's free-expression protection for political criticism must be respected even when hostile to minorities or dissenters.
[CJEU — verify Charter Arts. 11, 24; proportionality doctrine]

► EVP · Digital
Support for DSA enforcement, with a technical note. Algorithmic amplification is the multiplier on hate speech — the same comment gets different reach depending on the algorithmic system. The DSA's risk-mitigation requirements must address the algorithmic layer, not just content removal. Ensure the algorithmic-audit requirements under Article 34 are enforced robustly.

► Commissioner · Home Affairs
Support, operational. Online hate speech often targets migrants, asylum seekers and Roma communities — groups my portfolio protects. The DSA's systemic-risk framework should explicitly include hate speech targeting vulnerable populations. Coordinate with national law enforcement on the criminal investigation of hate-speech campaigns, particularly those organised across borders via Europol.

► Commissioner · Internal Market
No objection to hate-speech enforcement. The DSA applies uniformly across VLOPs; a targeted hate-speech law would create 27 different national regimes and fragment the single market. Keep enforcement at the VLOP level via DSA.

► Commissioner · Education & Culture
Support. Hate speech and media literacy are linked — young people need to recognise disinformation and hateful narratives. The DSA's transparency requirements (moderation reports, content policies) can be educational resources. A targeted hate-speech law is unnecessary; enforce the DSA and invest in media literacy.

► Commissioner · Research & Innovation
Support. Research on the link between hate speech and youth disengagement from democracy should inform the DSA's systemic-risk assessments. The European research community can supply evidence to guide proportionate moderation.

► President
I read the room. There is clear majority on DSA enforcement as the primary tool. There is consensus on rejecting targeted EU hate-speech legislation as legally and politically impracticable. The tensions: national definition fragmentation versus EU harmonisation — hate speech is not uniformly defined; an Article 83(1) proposal would require unanimity and would fail given Hungary, Poland and others' divergence. Better: DSA plus national prosecution. Content removal versus free expression — Justice's proportionality boundary: hate speech targeting minors is enforceable; political speech even when hostile to minorities must be protected. The DSA's requirement for reasoned removal decisions is the safeguard. Algorithmic amplification as the multiplier — the DSA's systemic-risk framework must address algorithmic ranking and recommendation, or we miss the harm vector. No blocking objection. Justice and EVP Democracy are firm that a targeted EU hate-speech law would stall in the Council. DSA plus national law is the practical path. I record consensus on a clear College line: strengthen DSA enforcement, reject targeted legislation, invest in media literacy. No formal vote.
[review — political judgement required on member-state acceptance of EU indicator list]

Deliberation closed · 9 positions recorded