Within hours of a federal officer shooting and killing a 37-year-old Minneapolis woman on Wednesday, some social media users identified two separate and unrelated people as the shooter, a move that for hours sent threats and vitriol their way.
Online posts claiming that “Steve Grove from Minneapolis” was the shooter began circulating Wednesday afternoon and quickly caught the attention of the publisher and chief executive of The Minnesota Star Tribune, whose name is Steve Grove. The paper he runs was in the throes of covering the news roiling the state, and the name of the Immigration and Customs Enforcement officer who had shot the woman was not yet public.
At first Mr. Grove said he was worried that it was a “horrible coincidence” — did he share a name with the shooter? But as time went on, he wrote in a Substack post and later told The Times, he realized something far more troubling was happening.
The posts “started piling up, and they got more and more viral,” Mr. Grove said. “It was clear this was like a vigilante hunt.”
About 600 miles to the south, in Springfield, Mo., a gun-shop owner named Steven Grove was also caught up in the fray. He said he received several Facebook messages Wednesday evening calling him the shooter, and on Thursday, his business received dozens of phone calls, many accusing him of murder. “It was surreal,” he said. Mr. Grove, a military veteran, said he has never been to Minnesota.
On Wednesday, an ICE officer shot and killed Renee Nicole Good in the driver’s seat of her car on a neighborhood street in Minneapolis. President Trump and other federal officials said the agent, identified Thursday as Jonathan Ross, was acting in self-defense. Others see it differently and state and local leaders are pressing for a state-run investigation.
What led to the erroneous information that the shooter was named Steve Grove is not entirely clear. Videos of the moment Ms. Good was shot show a masked agent firing into the vehicle Ms. Good was driving. By 2:49 p.m. on Wednesday, an image that appears to have been generated by Grok, the X chatbot created by Elon Musk, surfaced and claimed to show the officer’s full face. A user commented that the person was Steve Grove, who they said also works for The Minnesota Star Tribune.
The newspaper publisher does not bear a clear resemblance to the ICE agent. Nonetheless, that name went viral and caused mayhem for at least two Mr. Groves. About 6,200 posts mentioning Steve Grove have appeared across seven social media platforms, including X and Reddit, since Wednesday, said Jonathan Sperber, the chief operating officer of Gudea, a company that tracks online misinformation. About half of those posts specifically point to the Steve Grove in Minneapolis, he added on Saturday.
Mr. Grove, the newspaper publisher, said, “It’s just such a stark reminder of this world we live in, in which people are just searching for their own justice and not considering that maybe going to local news organizations to get valuable fact-checked information might be a good place to start.”
Amateur detectives on social media have long tried to find people using grainy photos, misreported names and other unreliable bread crumbs, often incorrectly identifying and then harassing people.
Days after a shooting at Brown University in December, for example, when the shooter was still at large, a Palestinian student became the baseless target of what one expert called “an internet witch hunt” after his photo was posted on X. And soon after Charlie Kirk was assassinated on Sept. 10, an X account with millions of followers used A.I. to turn photos of a person of interest into videos of him walking up a stairwell.
The suite of available generative A.I. tools is large and growing. Many of them have a tendency to make things up, a phenomenon known as hallucination.
“The problem, of course, is everybody wants their news and information fast,” said Hany Farid, a professor of computer science at the University of California, Berkeley, and a co-founder of GetReal Security. “But there’s a fast-accuracy trade-off: The faster you move, the more mistakes you make.”
Mr. Farid ran an experiment where he plugged an image of himself masked into three separate A.I. tools — Google’s Gemini, Grok and OpenAI’s ChatGPT — and asked them to show his uncovered face. Each of the resulting images looked like a different person.
Georgia Gee contributed research.
Pooja Salhotra covers breaking news across the United States.
The post Disinformation in Minneapolis Shooting Points at People Who Were Not Involved appeared first on New York Times.
