The Impact of AI-Generated Media on Public Perception of Alex Pretti’s Shooting

In the wake of the tragic shooting of Alex Pretti by federal officers in Minneapolis, AI-manipulated images and videos have permeated various social media platforms such as Facebook, TikTok, Instagram, and X. Many of these altered medias portray Pretti’s last moments, sowing confusion about the actual events surrounding the shooting.

Unlike many artificial intelligence-generated videos and images known as deepfakes that are easily debunked due to their unrealistic nature, the images concerning Pretti’s final moments closely mimic reality. As a result, they blur the line between fact and fiction, misleading many who encounter them online. This situation has led to people falsely asserting that authentic images of Pretti are fabrications created by artificial intelligence.

One widely shared image, apparently manipulated through AI, depicts Pretti, an ICU nurse, falling forward while a federal officer points a gun at his back. This image, viewed over 9 million times on X, has sparked controversy and a community note indicating that it was altered by AI. A particularly troubling detail in the image is the depiction of an ICE officer without a head.

Notably, some individuals in influential positions, including Sen. Dick Durbin (D-Ill.), have inadvertently propagated these manipulated images. Durbin presented the edited photo during a Senate speech, not realizing it was not an original image. Durbin’s spokesperson later conveyed regret for the oversight, emphasizing that the image used seemed authentic.

Aside from still images, various AI-generated videos have emerged online, including a TikTok clip illustrating a conversation between Pretti and an ICE officer, and a Facebook video displaying a police officer allegedly firing Pretti’s firearm. The latter post, flagged with a community note informing users of its AI enhancement, has amassed over 44 million views.

Ben Colman, co-founder and CEO of Reality Defender—a company specializing in detecting deepfakes—acknowledged that while AI-generated depictions of Pretti’s shooting are concerning, they are not unexpected. “The proliferation of grainy photos enhanced by AI has surged on social media lately,” Colman told NBC News. “These deepfakes, though claimed to clarify reality, tend to be poor approximations or outright fabrications.”

“Details like the absent head of a figure in some photos demonstrate the potential harm these fabricated images can cause when they gain viral traction,” Colman added. “Believing falsified images derails focus from seeking the truth.”

The widespread dissemination of AI-modified content has misled the public into suspecting the authenticity of genuine footage. Experts caution against this phenomenon, known as the liar’s dividend, where the falsely accused utilize AI manipulation claims to evade responsibility.

Three videos independently authenticated by NBC News depict Pretti in a confrontation with immigration agents days before his death. Yet, one video, which some social media users mistakenly deem AI-generated, shows Pretti kicking a federal vehicle before being subdued. Despite the video being verified by Pretti’s family and witnesses, some dismiss it as inauthentic.

Unfortunately, tools that decisively distinguish AI-fabricated from real content remain scarce. On X, the platform’s AI utility, Grok, offered responses that incorrectly labeled authentic footage as potentially AI-altered.

The proliferation of AI deepfakes has led to escalating misinformation about Pretti’s case, with misidentification adding to the disarray. As artificial intelligence technology advances, its capacity to foster realistic yet deceptive media increases.

In the past year, surges of AI-driven misinformation during pivotal news moments have become frequent, underscoring the necessity for improved awareness and vigilance among media consumers.

This analysis is brought to you by Jared Perlo, a fellow specializing in AI coverage with support from the Tarbell Center for AI Journalism. Contributions from Marin Scott, an Associate Reporter with the Social Newsgathering team, and Bruna Horvath, an intern on NBC News’ tech team, have enriched this exploration. Additional insights were provided by Colin Sheeley and Jean Lee.