Online Threats and Violent Speech – Know Your Rights

Threats of violence and calls for harm are among the most serious violations that can occur on social media – yet platforms routinely fail to act on them, or conversely remove legitimate content by mistake. Whether you were threatened in content on social media and it was not taken down after reporting, or your own content was removed under a platform's violence policy without justification, you have the right to appeal the platform decision under the Digital Services Act (DSA) by submitting a case with User Rights.

Since the Digital Services Act (DSA) came into force, users can challenge unfair moderation decisions through a certified out-of-court dispute settlement body. User Rights is the first such body certified under Article 21 DSA in Europe for Social Media platforms, certified by the German Federal Network Agency (Bundesnetzagentur). You do not need to complete the platform's internal complaint procedure before turning to User Rights. All we require is proof that you submitted a report or filed an internal appeal through the platform's complaint mechanism. Cases are currently reviewed for Instagram, Facebook, TikTok, LinkedIn, and Pinterest. Our assessments are based on platform policies and, where relevant, EU law as well as national law. We are certified for German and Italian law in applicable cases.

What Are Online Threats and Violent Speech?

Online threats are statements or visuals that threaten individuals or groups by expressing the intention, inciting or calling for violence against a person or group. This can include: messages threatening violence, death, sexual assault directed or threats or depictions at an individual. Violent speech is a broader category covering content that calls for, glorifies or incites violence against identified or identifiable people, even without a direct personal threat.

The distinction matters in practice. A direct threat – 'I'll beat you up when I see you' – is clearly prohibited under every major platform's policies. Violent speech is more contextual: a post calling for others to attack a named individual, a comment celebrating an act of violence against someone, or content that frames a specific person as a legitimate target for harm all fall within this category, even if no explicit personal threat is made.

Example: A journalist reporting on organised crime receives repeated Instagram comments referencing where they live and saying they "deserve what is coming to them". No comment states an explicit threat. Meta's Violence and Incitement policy does cover veiled and coded threats, but only where a threat signal and a contextual signal are both present, and Meta assesses credibility partly by reference to the target's public visibility and the risk to their physical safety. Where the journalist reports the comments together with evidence of that risk and the report is dismissed without engagement, a reviewable moderation failure occurs. Individual posts are reviewable on their own. But where you are the target of a series of veiled threats, submit all of them together. The meaning of each comment often only becomes apparent in combination with the others, and a submission covering the full pattern allows that assessment to be made.

Forms of Online Threats and Violent Speech covered by Platform Policies

Direct threats of violence, death or sexual assault against a named person
Messages referencing a person's address, workplace or daily routine alongside threatening language
Calls for others to attack, harm or 'deal with' a specific individual
Content glorifying or celebrating violence against an identifiable person
Threats conditioned on a future event ('if you publish that, you will regret it')
Coordinated campaigns that frame a person as a legitimate target for harm
Images or videos edited to depict violence against a specific person

What is not covered: context and legitimate expression

Not every reference to violence constitutes a prohibited threat. News reporting, documentary content, historical analysis, fiction and artistic expression involving violence are generally permitted. Hyperbolic language in everyday use, such as saying 'I could kill for a coffee right now' does not meet the threshold for a credible threat. The key factors are whether the statement targets a specific person and whether a reasonable person in that person's position would interpret it as a genuine announcement of harm.

Threats Policies on Social Media Platforms

All platforms covered by User Rights are subject to the DSA, which requires them to give reasons for moderation decisions and to cooperate with certified dispute settlement bodies. Each platform explicitly prohibits threats of violence and incitement to harm in its community standards.

The platforms' enforcement of these rules is often heavily reliant on automated systems that can both under-enforce (missing genuine threats) and over-enforce (removing legitimate content about violence in a news or educational context).

Meta: Instagram and Facebook

Meta's Violence and Incitement policy prohibits threats of violence, calls for violence against a target and statements of intent to commit violence. It also covers veiled or coded threats, but only where a threat signal and a contextual signal are both present. Threatening behaviour that falls short of incitement is dealt with under Bullying and Harassment. A single incident can engage both.

TikTok

TikTok's Community Guidelines on Safety and Civility prohibit threats of violence against individuals or groups, content that facilitates real-world violence, the glorification of violent acts against specific people, and instructions on how to commit harmful acts. Inciting or threatening serious violence where there is a credible risk of harm or danger to public safety can lead to account-level enforcement, and TikTok may also restrict or ban accounts for egregious off-platform behaviour.

LinkedIn's Professional Community Policies prohibit threatening, intimidating or harassing others, as well as inciting violence or calling for harm against specific individuals. Because LinkedIn runs on verified professional identity, threats here often appear as implied consequences to someone's employment or standing rather than as explicit violent language. That makes them harder for automated review to detect. Implied consequences and coordinated pressure campaigns all fall within the scope of the policy, even without explicit violent language.

Pinterest's Community Guidelines addresses online threats in their policy on violence and threats. Pinterest states it is not a place for graphic violence or threatening language and removes or limits the distribution of such content, including content showing the use of violence, threats and language that glorifies violence. Threatening behavior can occur in comment sections or through content created specifically to intimidate a named person.

What Can You Do if You Are Affected?

There are two situations in which User Rights can help: the platform has not acted on threats or violent speech you reported, or your own content has been removed under a platform's violence policy without justification.

Scenario A: You received threats and the platform has not acted

Secure the evidence immediately: Screenshot the threatening content with the profile name, URL, date and time before reporting it. Save screenshots to a location outside the platform. If the threat arises from multiple pieces combined, submit them to User Rights in one case. Otherwise, User Rights only assesses the single piece of content, and might be unable to find a violation where one actually occurred.
Report to the platform: Use the in-built reporting function. Be specific: describe why the content constitutes a threat rather than merely offensive language, and reference the exact wording. If there is a pattern of threatening messages, indicate this in your report. Also reference all relevant links in your complaint so the platform gets a cohesive picture of your situation, if possible.
Appeal if dismissed: If the platform does not act on your report, use the appeal function available via the help centre or the dismissal notification. Keep a record of the appeal and the platform's response. Note that you do not have to wait for a response by the platform. If your complaint remains unanswered after a reasonable time of seven days, you can file a case with User Rights if you submit proof of your submission.
Submit to User Rights: If the platform's inaction was inconsistent with its own threats or violence policy, submit a complaint to User Rights via our online form. The review is free of charge for users.
Consider reporting to the police in parallel: For serious or credible threats, reporting to law enforcement is advisable regardless of the platform moderation outcome. Platform moderation and the User Rights procedure do not replace criminal law remedies.

Scenario B: Your own content was removed under a violence policy

Check the reason: Under Art. 17 DSA, platforms must give a specific reason for every moderation action. Verify whether the reason cited actually applies to your content – for example, whether journalistic, artistic or educational content was misclassified as incitement.
Appeal internally: Use the platform's appeal mechanism. Provide context that makes clear the legitimate purpose of your content – for example, that it documents a threat you received rather than making one.
Submit to User Rights: If the appeal was unsuccessful or unanswered, submit your case via our online form. User Rights will assess whether the removal was consistent with the platform's own policies and applicable law.

Note: User Rights does not review the content of direct messages, or moderation actions that were taken more than six months ago. A full overview of what falls within scope is available on the scope of review page.

Has a platform failed to act on threats you reported – or was your content wrongly removed under a violence policy?

Submit your case to User Rights. Our certified dispute settlement procedure is free of charge for users.

SUBMIT A CASE

Frequently Asked Questions (FAQ)

A direct threat is a statement addressed to a specific person announcing harm to them. Violent speech is broader: it includes calls for others to harm someone, content that glorifies violence against a named individual, or material that frames a specific person as a legitimate target. Both are prohibited under platform policies, and both are reviewable by User Rights.

No. User Rights reviews both directions: inaction on reported threats, and the wrongful removal of content under a violence policy. If your post documenting threats you received – or legitimate journalism about violent incidents – was removed, that decision is reviewable.

Threats focus on the announcement of specific harm to a person. Cyberbullying is a broader pattern of repeated degradation, humiliation or harassment that may or may not include explicit threats. A single threatening message can constitute a reviewable incident on its own; cyberbullying requires a pattern. The two frequently overlap.

User Rights reviews moderation decisions on public content: posts, comments, stories, videos, live-videos, and public profiles. Threats sent exclusively via direct messages fall outside the scope of review. If the same person sent both public threatening content and private messages, the public content is reviewable.

The procedure is free of charge for users. The platform bears the costs of the dispute settlement process.