Violence and Incitement

Policy Rationale

Our Transparency Policy is deeply committed to fostering a safe and respectful environment for conversation and interaction. Central to this commitment is our stance against violence and incitement. We recognize the potential for content on our platforms to influence offline behavior, and we take proactive measures to mitigate the risk of real-world violence.

In understanding that expressions of disdain or disagreement may sometimes manifest through threats or calls for violence, we maintain a strict policy to remove any language that incites or facilitates violence and credible threats to public or personal safety. Our focus is on preventing content that deters constructive conversation and poses a threat to individuals or groups, irrespective of the specific characteristics or status of those involved.

Context plays a crucial role in our evaluation process. We carefully consider various factors, including whether the content is raising awareness of violence, condemning violent threats, or discussing non-credible threats in a manner that does not contribute to a climate of fear. Our aim is to distinguish between content that poses a genuine risk and that which contributes to meaningful dialogue.

In instances where content is deemed to pose a genuine risk of physical harm or direct threats to public safety, we do not hesitate to remove such content, disable accounts, and collaborate with law enforcement authorities. Our actions reflect our dedication to maintaining a platform where open, respectful, and safe conversations can flourish, free from the shadow of violence and incitement.

We Remove:

We actively prevent the dissemination of violence and incitement on our platform by removing content that poses threats of violence in various forms. These threats can manifest as statements or visuals that express an intention, desire, or call for harm against individuals or groups, and may include statements of intent, calls to action, advocacy, aspirational statements, and conditional threats.

However, our policies are designed to distinguish between harmful content and expressions shared in contexts that aim to raise awareness, condemn violence, or discuss less severe threats within the realm of contact sports, or against recognized violent entities.

Universal Protections:

Our platform offers universal protections, safeguarding all users from:

• Threats of violence that pose a risk of death or severe harm.

• Threats that could result in serious injury. These are removed when they target public figures or groups not identified by protected characteristics if deemed credible, and against any other targets, including those identified by protected characteristics, regardless of credibility.

• Admissions of severe violence, except in contexts of redemption, self-defense, sanctioned contact sports, or when actions are carried out by law enforcement or military personnel.

• Threats or depictions of kidnapping or abduction, with exceptions for content shared by victims or for purposes of awareness, condemnation, or seeking help.

Additional Protections:

Specific groups, including private adults, children, high-risk individuals, and those targeted based on protected characteristics, receive additional protections against low-severity violence threats.

Other Forms of Violence:

We also remove content related to:

• Solicitation or admission of services for severe violence, such as hiring hitmen or advocating for harmful practices.

• Instructions on creating or using weapons intended to cause serious harm, except in contexts like recreational self-defense, military training, certain video games, or legitimate news coverage.

• Guidelines on making or using explosives, barring non-violent purposes such as education, commercial games, or scientific research.

• Threats involving weapons or forcible entry into sensitive locations, including but not limited to places of worship, educational institutions, or during elections.

• Violence-related threats concerning voting, registration, election administration, or outcomes, irrespective of a specific target.

Our commitment to these guidelines ensures that while we champion free expression and meaningful dialogue, we do not compromise on safety or the integrity of our community.

For certain Community Standards, additional information and context are crucial for our enforcement actions. These guidelines help us address complex situations where threats may not be explicitly stated but are implied through coded language or context. Our aim is to carefully evaluate and act on content that poses a real threat to safety while maintaining the integrity of open dialogue.

Areas Requiring Further Review:

We Review and May Remove:

• Threats Against Officials: Any threats directed at law enforcement officers or election officials, irrespective of the public figure status or the threat's credibility.

• Coded or Veiled Threats: Statements where violence is implied rather than explicitly stated, identified through a combination of a threat signal and a contextual signal, as outlined below:

o  Retaliatory Threats: Expressions wishing for violence in retaliation to a real, perceived, or anticipated grievance.

o Historical or Fictional Violence References: Threats that invoke known incidents of violence from history or fiction as a means to threaten.

o Threatening Calls to Action: Invitations or encouragements for others to commit violent acts or to participate in such acts.

o Sensitive Information Exposure: Sharing or implying knowledge of personal information that could escalate the credibility of a threat, such as someone's home address, workplace, educational institution, daily routes, or current location.

Contextual Considerations:

• Imminent Violence Indicators: Local context or expertise indicating that a statement could lead to imminent violence.

• Reports by Targets: Content reported to us by the target or an authorized representative.

• Implicit Threats to Bring Weapons: Indirect threats to bring weapons to sensitive locations (including places of worship, educational facilities, polling places, or other sites during heightened risk of violence) or encouragements for others to do the same.

Election-Related Concerns:

• Violence-Linked Claims: Assertions or speculation about election-related corruption, irregularities, or bias, especially when paired with indicators of violent intent (e.g., threats involving weapons, visual depictions of weapons, references to arson, theft, vandalism).

o      Targeting individuals or locations, regardless of specificity.

o      Where the target is not explicitly mentioned.

• Election-Related Gatherings: References to election-related events or gatherings, accompanied by indicators of violent intent.

Our approach to these areas is guided by a commitment to safety and the promotion of a respectful, open dialogue. We recognize the importance of nuanced review in these contexts to ensure our platform remains a safe space for all users.