Indiana livestreamed shooting proliferates on X and Facebook

3 July 2024

By: Moustafa Ayad and Katherine Keneally


In May, a 20-year-old man livestreamed an attempted mass shooting at a grocery store in Indiana. Although there were no victims, the video – which featured multiple shots aimed at Black shoppers and employees while the attacker vocally espoused racial slurs – was inarguably violent, graphic and disturbing. Despite this, ISD identified dozens of versions of the full video on X and Facebook, generating nearly 9 million views in the 48 hours after the incident took place. Several of the videos are still available on both platforms, and have generated subsequently more views. ISD researchers also searched for the video on Instagram, TikTok and YouTube, but found only news content and snippets of the video without the depiction of violence.

The video appears to have been shared with ease across both platforms. No content warning labels were assigned, though the video likely violated both X and Facebook policies (explored below). The unchecked proliferation of this video suggests that policies surrounding violent content on these platforms are not being consistently enforced.

It is not clear if the event led to the Global Internet Forum to Counter Terrorism (GIFCT) Content Incident Protocols (CIP) being activated in response to the livestreamed shooting, likely because the event had no clear ideological motivation. The CIP is designed to respond to the dissemination of terrorist or violent extremist events and incidents online, and to stop the spread of performative, livestreamed ideological violence. No motivation for the shooting beyond the attacker’s expressed desire to simply “kill somebody” has been established to date. 

  • ISD found 30 full versions of the livestreamed attempted shooting on X which generated more than 8.9 million views. The platform, which recently updated its guidelines for violent content and speech, asks its users to voluntarily change their content setting if they regularly post graphic media (noting that their settings will be changed for them if they continue to post graphic media and do not change the setting themselves), so that a content warning will be automatically applied to all posts from the account. Users also have the option to apply a “one-time content warning” on individual graphic posts. While graphic media that includes “violent crimes” is permitted when appropriately labelled, the content in this instance appears to fall under X’s definitions of both “wish of harm” and “violent threats”, both of which are strictly prohibited. More broadly the violent content policy indicates “explicitly threatening, inciting, glorifying, or expressing desire for violence is not allowed.” It is unclear if the platform considers the suspect’s statement, “I’m going to shoot 11 people,” to meet the threshold for removal under its violent content policy.
  • Just over half of the full versions of the livestreamed video shared on X were posted by Premium users on the platform. 17 of the 30 videos — 53 percent — were shared by Premium X users, generating 8,901,020 views, well over 99 percent of all views of the video. This indicates that Premium users received the most traction on their posts of the video. The 13 users without the Premium status on X that posted the same video generated just 27,306 views.
  • Analysts found 13 full versions of the livestreamed attempted shooting on Facebook, generating some 29,891 views. Four of the 13 videos did not display view or play metrics, therefore it is unclear how many more views were generated on the platform. None of the videos came with platform-applied graphic content labels.
  • A review of Klaff Jr.’s Facebook page did not show any ideological grievances or racial animosity. Users that learned of the shooting took to Klaff Jr.’s Facebook page to admonish him for being a racist. Some followers of the personal page apparently set up by Klaff Jr., which had been live since July 2011, defended him against these claims, based on their interactions with him offline.
  • It is not readily apparent the GIFCT’s CIP was activated in response to the livestreamed shooting, likely because the event had no overt ideological motivation. The GIFCT protocols have played a vital role in blunting the effects of livestreamed violent extremist and terrorist events in the past. In 2023, the GIFCT activated the protocols twice, according to its annual transparency report for that year. First, for a livestreamed shooting in Kentucky, which resulted in five deaths and 8 injuries, and second, in response to the October 7 Hamas attacks. For incidents that do not clearly or immediately meet the threshold for a response by GIFCT or similar bodies/organisations, there may be a need for better informal cross-platform coordination, or for GIFCT to be able to still share information in cases where an incident is assessed, but deemed not to meet their criteria for T/VE incidents and therefore not trigger a formal response.    

“Going out with a bang”

The broadcast on 23 May 2024 by Klaff Jr., shows the suspect entering a grocery store in Fort Wayne, Indiana. Klaff Jr. began a Facebook livestream where he is seen inside a bathroom while brandishing a .40 caliber handgun.

He looked directly into his phone’s camera and stated: “I’m going to shoot 11 people.”

He then lit a Black and Mild cigar and asked people to watch and screen record the livestream. He subsequently began filming as he walked around the store asking viewers to choose whom he would shoot. When someone suggested he shoot an elderly man, he said, “I ain’t going to kill no old ass man.”

Klaff Jr. apparently told his livestream he was “going out with a bang” before he pinpointed a “target” near the bakery inside of the store and fired six shots. During the livestream, Klaff Jr. used a racial slur for Black people several times and began firing at a Black woman at the bakery. Luckily, no one was injured. A review of Klaff Jr.’s public social media footprint did not seem to indicate that he was linked to any violent extremist or terrorist groups.

Klaff Jr., who was apprehended hiding near a Dollar Store dumpster after buying a replacement t-shirt, reportedly told police he “always wanted to kill somebody, that’s all I wanted to do. I was supposed to go to the military. I couldn’t wait on that, obviously.” According to a police affidavit, Klaff Jr. was choosing victims in the store because they “were standing in a position with no cover.”

An attempted mass shooting goes viral

In the aftermath of the shooting, the livestream of the incident began proliferating on X (formerly Twitter) and Facebook. ISD analysts found 43 versions of the full livestreamed video on X and Facebook in the 48 hours after the incident took place, which generated more than 8.9 million views and plays (Facebook has recently changed its viewing metrics from views to plays). None of the full livestreamed videos on either platform came with a platform-generated content warning.

In May, X also changed its violent speech and content policy, providing guidance for users who post violent content, though the policy notes that “threatening, inciting, glorifying, or expressing desire for violence is not allowed.” Facebook specifically prohibits content that promotes “threats of violence that could lead to death (or other forms of high-severity violence)” as per its violence and incitement policy.

Based on previous livestreamed shooting incidents where a CIP was activated, Klaff Jr.’s actions and livestream could fit some of the criteria for activating the protocols, but not all, posing a potential gap in cross-platform coordination. These protocols were enacted following the Christchurch mass shooting in which 51 people were killed and 40 others injured. However, the four-point CIP criteria as outlined by GIFCT presents some limitations around its activation for this particular incident.  While the incident meets the first criteria of the CIP as an attempt to cause “mass violence” it does not appear to be “terrorist” or “violent extremist” in nature at this point in time. The incident does meet three other criteria noted within the protocol: it was livestreamed, depicted attempted murder, and was actively distributed on GIFCT member platforms.

Conclusion

The spread of the Fort Wayne livestream video across X and Facebook illustrates potential limitations to the accurate and timely enforcement of the platforms’ own policies. X and Facebook do not appear to have followed their own stated platform policies on violence, as the video generated more than 9 million views, and continued to proliferate after the initial 48 hours of its posting.

Though the physical impacts of the attack could have been much worse, the psychological impacts on both those targeted in the incident and individuals who have watched the video remain significant. Though studies have shown that viewing violent screen content can have significant impacts on mental health and behavior, particularly for youth, this damage is difficult to measure, and the impacts of this graphic and violent video are likely yet to be seen.

False and unverified claims proliferate online following Trump assassination attempt

Unverified claims about the attempted assassination of former President Donald Trump proliferated across social media in the 24 hours following the incident and continue to spread in the absence of more detailed factual information about the shooter. ISD identified the main false claims being promoted and how they mutated and spread across platforms.