What actually happened
On 4 November 2026 the European Commission adopted its 2027 work programme. According to Brussels Signal, an annex lists the December 2021 hate speech proposal among files the Commission "intends to drop within six months." The precise status matters. This is a stated plan to withdraw, not a completed withdrawal. Reports so far rest on secondary outlets, and we have not found a Commission press release announcing it.
The 2021 initiative was not a bill that criminalised anything directly. On 9 December 2021 the Commission proposed adding hate speech and hate crime to the list of "EU crimes" in Article 83(1) of the Treaty on the Functioning of the European Union. That list covers offences such as terrorism and trafficking, where the EU may set minimum criminal-law rules. The Commission's own release says the Council must adopt the extension unanimously, after Parliament consents. Only then could it table an actual legislative proposal. That first step never cleared the Council.
The strongest case for the original plan
Supporters have a serious argument. The EU's 2008 Framework Decision already requires member states to criminalise public incitement to violence or hatred on grounds such as race, colour, religion, descent or national or ethnic origin. The Commission's 2021 release says so. National implementation, however, is uneven. A victim's protection can depend on where the offender sits. A European Parliamentary Research Service briefing noted that Parliament "enthusiastically welcomed" the proposal while member states in Council "have been unable to agree." Harmonised definitions across 27 jurisdictions would also give platforms one rulebook instead of many. That is a real benefit for a single market.
Why the stall is not a catastrophe
Unanimity is a high bar by design. Criminal law is the sharpest tool a state has against speech, and the treaty deliberately makes governments agree before the EU gets new competences. Where 27 governments with very different speech traditions cannot agree, the legitimate reading is that there is no consensus on where the line sits, not that a procedural trick has defeated the public interest.
The Commission had also offered no hard definition of "hate speech" beyond the existing incitement standard. Expanding the EU crime list to a category whose boundaries differ between Warsaw, Paris and Dublin invited the over-criminalisation that chills lawful debate. The existing floor already covers the clearest harms: incitement to violence or hatred. The case for going further was never tied to evidence that the floor was failing in ways harmonisation would fix.
The online-only directive is the real question
Brussels Signal reports that the Commission is instead weighing a directive on online "incitement to violence and hatred," which it says would need only a qualified majority rather than unanimity. We treat that as reported, not confirmed. If accurate, the shift changes the politics. A lower voting threshold means fewer governments can block it, which is exactly why the drafting would matter more.
On 9 November, Alliance Defending Freedom International, as quoted by ZeroHedge, called a possible directive "yet another tool to silence disfavored views and strengthen the [Digital Services Act's] online censorship regime." That language is rhetorical, and the directive does not exist yet. But the underlying worry is not frivolous. Online-specific criminal rules interact with platform liability. If a platform risks regulatory exposure for hosting content that a national court might later call unlawful, the rational response is to remove borderline speech early. The Digital Services Act already pushes in that direction through notice-and-action duties, and ZeroHedge, summarising reporting, cites January 2025 DSA guidelines on 24-hour review of notices. Layering a new criminal-law definition on top multiplies those incentives.
What a proportionate directive would look like
A pro-innovation, pro-speech approach does not require rejecting every online incitement rule. It requires three design choices.
- Keep the incitement threshold narrow. Limit it to incitement to violence or hatred in the existing 2008 sense, with an intent requirement and a real likelihood of harm, rather than offence, insult or opinion.
- Target the speaker, not the intermediary. Criminal liability should attach to the person who incites. Platform duties should stay within the DSA's existing framework instead of creating a second, harsher liability track.
- Require evidence first. The Commission should publish data on prosecution gaps and cross-border enforcement failures before proposing new law. The 2021 plan lacked that grounding in its public materials.
The Commission's own 2021 materials cited a striking figure: 52% of young women and girls have experienced online violence. That points to a harm worth addressing, though it concerns online violence broadly and does not by itself show a gap that a new hate speech offence would close. Targeted measures against threats and harassment, which are narrower and easier to define, may do more than a broad speech category.
Bottom line
Withdrawing a proposal that could not win unanimity is a normal outcome of the treaty's design, and the five-year deadlock is a reasonable signal that the original framing was too broad. The risk now is that a procedurally easier route, qualified-majority voting on an online directive, lets through what member states declined to endorse. The Commission should show its evidence and publish a precise definition before asking Parliament and Council to weigh in. Until a text exists, critics and supporters alike are arguing about a draft nobody has read.