AI Error Jails Innocent Tennessee Grandmother for Nearly 6 Months
Key Facts
- What: Fargo Police used facial recognition software that wrongly identified Angela Lipps, a 50-year-old Tennessee grandmother who had never visited North Dakota, as the suspect in an organized bank fraud case.
- Duration: Lipps spent nearly six months in jail — 108 days in a Tennessee jail without bail followed by additional time in North Dakota’s Cass County Jail.
- Consequences: She lost her home, car, and dog; charges were dismissed on Christmas Eve 2025 after bank records proved she was 1,200 miles away in Tennessee committing routine transactions.
- Investigation Flaws: Police relied primarily on the facial recognition match, supplemented by a review of her social media and driver’s license photo, without interviewing her or verifying her location before arrest.
- Legal Outcome: All charges dismissed December 24, 2025; lawyer Jay Greenwood criticized over-reliance on the AI tool.
Lead paragraph
A Tennessee grandmother who had never left her home state was arrested at gunpoint, extradited nearly 1,200 miles, and jailed for almost six months after Fargo Police’s facial recognition software mistakenly identified her as the perpetrator of a bank fraud scheme in North Dakota. Angela Lipps, 50, lost her home, car, and dog while sitting in jail as a fugitive from justice she knew nothing about. The case, which relied heavily on the flawed AI match, was dismissed on Christmas Eve 2025 once her bank records proved she was buying pizza and depositing Social Security checks in Tennessee at the exact times police claimed she was withdrawing tens of thousands of dollars in Fargo.
How Facial Recognition Led to Wrongful Arrest
In April and May 2025, Fargo detectives were investigating multiple incidents in which a woman used a fake U.S. Army military identification card to withdraw large sums of money from banks. Surveillance video captured the suspect clearly.
Rather than relying solely on traditional investigative methods, Fargo police turned to facial recognition software to identify the woman. According to court documents obtained by WDAY News through an open records request, the software returned a match for Angela Lipps. A detective then examined Lipps’ Tennessee driver’s license photo and her social media accounts. In the charging documents, the detective stated that Lipps appeared to match the suspect based on facial features, body type, hairstyle, and hair color.
No one from the Fargo Police Department contacted Lipps by phone or attempted to verify her whereabouts before seeking charges. On July 14, a team of U.S. Marshals arrested Lipps at her home in north-central Tennessee while she was babysitting four young children. She was taken away at gunpoint.
“I’ve never been to North Dakota, I don’t know anyone from North Dakota,” Lipps told WDAY News. “It was so scary, I can still see it in my head, over and over again.”
Months Behind Bars With No Bail
Because she was charged as a fugitive from justice, Lipps was held without bail in a Tennessee jail for 108 days. She was not transported to North Dakota until Oct. 30. The next day she made her first court appearance in Fargo.
Lipps, a mother of three and grandmother of five who had spent nearly her entire life in Tennessee, had never flown on an airplane. Her travels had been limited to neighboring states.
In North Dakota she faced four counts of unauthorized use of personal identifying information and four counts of theft. Jay Greenwood, the attorney appointed to represent her, immediately requested her bank records from Tennessee.
On Dec. 19 — more than five months after her initial arrest — Fargo police finally interviewed Lipps for the first time at the Cass County Jail. The bank records showed she was depositing Social Security checks, buying cigarettes at a gas station, purchasing pizza, and using a cash app for Uber Eats in Tennessee at the precise times the fraud was occurring in Fargo.
Five days later, on Christmas Eve, prosecutors dismissed all charges and Lipps was released. She was left stranded in Fargo in summer clothes with no coat as snow covered the ground.
Lawyer Criticizes Over-Reliance on AI
Greenwood was blunt in his assessment of the investigation.
“If the only thing you have is facial recognition, I might want to dig a little deeper,” he told WDAY News.
The case highlights growing concerns about law enforcement’s increasing use of facial recognition technology without sufficient safeguards. Multiple studies and previous high-profile cases have shown the technology can produce false positives, particularly when matching individuals from different regions or demographic backgrounds.
This is not an isolated incident. The Innocence Project has documented other cases, including that of Porcha Woodruff, a pregnant Detroit woman arrested based on a facial recognition match. Critics argue that when AI tools are used as the primary basis for probable cause rather than as one investigative lead among many, the risk of wrongful arrests rises dramatically.
Devastating Personal Impact
Beyond the 170-plus days spent in jail, the human cost to Lipps has been severe. While incarcerated she lost her home, her car, and her dog. Upon release she had no immediate way to return to Tennessee and faced the daunting task of rebuilding her life from scratch.
The emotional toll remains visible. Lipps described the trauma of the armed arrest in her own home and the helplessness of being held for months for a crime she could not possibly have committed.
For the first time in her life, she found herself in a state she had never visited, facing charges based on technology she had no reason to distrust until it upended her world.
What This Means for Law Enforcement and AI Use
This case arrives as police departments across the United States expand their deployment of facial recognition tools. While the technology can accelerate investigations, the Lipps case illustrates the danger when it is treated as near-conclusive evidence rather than a starting point requiring rigorous corroboration.
Legal experts say the incident raises questions about training, policies, and accountability standards for departments using AI-driven identification systems. At minimum, best practices recommend that a facial recognition match trigger additional verification steps — such as alibis, location data, financial records, or direct interviews — before an arrest warrant is sought.
Fargo Police have not yet issued a detailed public statement on changes to their facial recognition protocols following this case. The department’s reliance on the software, followed by what appears to have been minimal follow-up investigation, has drawn criticism from both Lipps’ attorney and civil liberties advocates.
What’s Next
Lipps is now focused on rebuilding her life in Tennessee. Whether she will pursue civil action against the City of Fargo or the vendors of the facial recognition software remains unclear.
The case is likely to fuel legislative efforts in multiple states to regulate law enforcement use of facial recognition. Several jurisdictions have already banned or heavily restricted the technology in criminal investigations due to accuracy and bias concerns.
For developers and companies creating law enforcement AI tools, the incident serves as a reminder that technical accuracy in controlled tests does not always translate to real-world reliability when lives and liberty are at stake.
As facial recognition and other AI systems become more deeply embedded in policing, the Angela Lipps case stands as a stark example of how quickly technology can go wrong when human oversight and traditional investigative work are deprioritized.
Pull Quote
“If the only thing you have is facial recognition, I might want to dig a little deeper.” — Jay Greenwood, Angela Lipps’ North Dakota attorney
This case changes how both police departments and the public view the reliability of AI identification tools. What was sold as a powerful crime-fighting technology has, in this instance, destroyed an innocent woman’s life for nearly half a year.