March 28, 2024

Shaheed | Meta’s Oversight Board Loosens Policy Regulating Glorification of Violence

CyberWell warns that the recent opinion released by Meta’s Oversight Board loosening restrictions on use of the term “shaheed” may lead to harmful consequences and promote pro-terror and violent content online.

Following the recent opinion released by Meta’s Oversight Board regarding use of the term “shaheed” when referring to dangerous individuals or organizations, CyberWell warns that the loosening of this policy can lead to serious digital and real-world harm. With this safeguard removed, we urge Meta to direct additional resources to enforcing its policies against glorification of violence and pro-terror content in order to close the already existing enforcement gaps which allow this prohibited content to remain online.

Meta Requests Guidance on Removal of “Shaheed”

In February of 2023, Meta submitted a request to their independent Oversight Board to review their policy on removing the term “shaheed” in Arabic (typically translated as “martyr”) when used to refer to an individual affiliated with dangerous organizations or when praising violent actions committed by this individual. Over one year later, the Oversight Board has finally released its policy recommendation, finding “that Meta’s approach to moderating content that uses the term “shaheed” to refer to individuals designated as dangerous substantially and disproportionately restricts free expression.”

In 2023 CyberWell submitted a Policy Advisory Opinion on this issue – backed by data.

Our analysts vetted 300 pieces of content that were live on Facebook, flagged as highly likely to be antisemitic, and contained the word “shaheed”. Based on this dataset we determined that, not only was content containing the word “shaheed” and praising a dangerous individual or organization readily available on the Facebook platform despite violating Meta’s existing policy, but also, from a safety perspective, the existing policy did not go far enough.

CyberWell urged the Oversight Board to not only maintain its current policy of heavily monitoring “shaheed” when paired with DOIs (Dangerous Organizations & Individuals), but also to dedicate additional resources to enhance enforcement of the current policy and to implement additional recommendations for policy improvements to better identify and remove pro-terror and violent content at scale.

Leading up to the release of the Oversight Board’s opinion, CyberWell conducted a dive into Meta’s platforms to see if the term “shaheed”, when paired with DOIs, is being used to praise and glorify violence today.

“Shaheed” Continues to Glorify Terror Attacks on Meta Platforms

CyberWell affirms that “shaheed” is a nuanced term – it can refer to those killed young, unexpectedly, and in circumstances unrelated to terrorist attacks. Such content is legitimate and should not be removed. However, the term “shaheed” is also often used to praise or encourage acts of violence and terrorism in a way that absolutely violates Meta’s Community Standards as well as human decency.

Similar to the insights from our original Policy Advisory Opinion, our new dataset demonstrates that the negative context of the term “shaheed” continues to appear on Meta platforms and includes, among others, the following categories:

 

The examples below, coupled with the data that CyberWell collected and analyzed a year ago, clearly demonstrate a pattern of content – live on Meta platforms – using the term “shaheed” to encourage, support, justify, and glorify violence and hatred, particularly against the Jewish people.

While the Oversight Board feels that Meta’s other polices are enough to address any glorification of violence, as they state “these policies, enforced accurately, mitigate the dangers resulting from terrorist use of Meta’s platforms”, it is clear to see that these policies, in fact, are NOT enforced accurately or sufficiently. By removing the additional safety that this policy provided, the Oversight Board opens the door to looser regulations on calls to and glorification of violence, romanticization of terrorism, and the expansion of its supporter-base among social media users. Now that the Oversight Board has gutted this safety mechanism, CyberWell urges Meta to enforce and improve enforcement of its existing Community Standards against violence and hate at scale.

Examples | Existing Policy Enforcement Gaps

(Prior to the Oversight Board’s decision)

The examples below represent a pattern of failure to enforce existing policy. They each contain use of the term “shaheed” to refer to an individual who is known to be affiliated with a dangerous organization – even including a Hamas leader and founder of the al-Qassam Brigades military wing.

Shaheed + Reference to a Recognized DOI

Hamas

The post below uses the term “shaheed” to praise Ahmed Al-Ghandour, who was the commander of the Northern Gaza al-Qassam Brigades – the military wing of Hamas.

The post below uses the term “shaheed” to praise Saleh al-Arouri, who was a senior leader of Hamas and a founding commander of its military wing, the al-Qassam Brigades.

Hezbollah

The post below uses the term “shaheed” to praise Hezbollah terrorists.

This template is frequently used to glorify Hezbollah terrorists and should be monitored by Meta’s content moderation team on an ongoing basis. It includes:

 

The group below uses the term “shaheed” to commemorate Hezbollah terrorists.

Followers: 3.5K

Examples | Existing Policy Gaps

Shaheed + Reference to Organizations and Individuals NOT Designated as FTOs but which Engage in Serious Offline Harm

CyberWell identified a number of posts praising individuals who committed violent terrorist attacks but who were not associated with a recognized terrorist organization. It is important to note that Meta uses the US State Department’s list of Foreign Terrorist Organizations (FTOs), which is far from comprehensive due to the very nature of terrorist groups, particularly in volatile regions. Meta’s policy only extends to individuals affiliated with recognized terrorist organizations and does not extend to unaffiliated individuals who carry out independent violent acts.

The post below uses “shaheed” to praise an individual who carried out a violent attack but who was not affiliated with a recognized FTO.

Caption: The group of the mujahid shaheed Ibrahim al-Nabulsi

NOTE: Ibrahim al-Nabulsi was associated with the Lion’s Den, an organization known to carry out violent terrorist attacks but not yet recognized by the US State Department as a Foreign Terrorist Organization.

The post below uses “shaheed” to praise an independent act of violence committed against Israeli civilians.

On March 14, 2024, Fadi Abu Latif went to a gas station near Bet Kama (Southern Israel), attacked 3 Israeli civilians in a cafe with a knife, killing one, and was himself killed. This post celebrates his violent attack and glorifies acts of murder.

Co-opting Terror

Since the events of October 7 CyberWell has seen a significant spike in the glorification of violence in both Arabic and English. In Arabic alone, whereas prior to October 7, 75.6% of antisemitic content vetted by CyberWell promoted classic stereotypes, tropes, and conspiracies against Jews, post-October 7 saw 61.2% of antisemitic content calling for or justifying violence against Jews.

Highly worrying is the romanticization of terrorist acts, revisionist history to glorify and justify violence, and appropriation of terrorist terminology. This became apparent when Osama bin Laden’s Letter to America began making the rounds on TikTok in November, just a month after the October 7 attack. Young social media users created videos expressing that they identified and empathized with his anger and objections, much to public dismay. Recently, CyberWell further noted several instances where the English term “martyr” was used to praise violent actions, possibly denoting an adoption of “shaheed” into English pro-terror content.

CyberWell Rejects Violence – and so should Meta

When social media platforms allow harmful content promoting and glorifying violence to remain online, they are complicit in the real-world repercussions.

Meta’s Oversight Board acknowledges that “acts of terrorist violence have severe consequences – destroying the lives of innocent people, impeding human rights and undermining the fabric of our societies” and the recommendation further goes on to say that “the Board’s recommendations start from the perspective that it is imperative Meta take effective action to ensure its platforms are not used to incite acts of violence, or to recruit people to engage in them”. However, CyberWell remains deeply skeptical that this policy recommendation is the correct step. Particularly at this moment when tensions are at an all-time high, the rates of antisemitic incidences continue to skyrocket, and Jewish communities around the world fear for their lives, now is not the time to loosen safety standards and hope for the best.

CyberWell will continue working with social media platforms to identify enforcement and policy gaps and make data-backed recommendations to create safer digital spaces for Jewish users.  We urge the platforms to remove glorification of violence at scale.  And we call for social media users and society at large to choose a better path – reject violence and hate and work toward a better future for all.

More Articles
January 9, 2025
New Orleans Terror Attack Spawns Antisemitic Conspiracy Theories

Antisemites wasted no time in blaming Jewish people for the New Orleans terror attack, turning the horrific incident into an anti-Jewish, conspiracy-fueled hate fest.

Logo for CyberWell, the first-ever open database of online antisemitism
January 7, 2025
CyberWell Statement on Meta’s Announcement to End Fact Checking, Shift to Community Notes

This is not a victory for free speech – it’s a systematic lowering of the bar that means less active enforcement from Meta against antisemitism and hate speech.

Logo for CyberWell, the first-ever open database of online antisemitism
November 15, 2024
Press Release | Election Related Antisemitic Social Media Posts Removed 30% Less Than Entire 2023

Our findings highlight how election-related discourse on social media has intensified the spread of dangerous conspiracy theories about Jews, with antisemitism emerging as a prominent feature in the political dialogue surrounding the U.S. elections on both sides of the political spectrum.

Contact Us

Fill out this form with some details or email us at [email protected]

Be in touch to request a platform demo, learn about our
work, explore partnership opportunities, offer support, or
simply to encourage our efforts. We want to hear from you!