AI Misidentification Leads to Wrongful Arrest of Tennessee Grandmother in North Dakota Fraud Case
March 29, 2026
The Fargo Police Department acknowledged errors in the case, including reliance on a neighboring agency's AI system that led to Lipps being identified as a suspect and arrested in Tennessee before being extradited to North Dakota.
Lipps has stated she will never return to North Dakota, stressing how the ordeal has affected her liberty and reputation.
Lipps’s lawyers call the detention traumatic and are pursuing civil rights claims, criticizing the initial investigation for treating AI as a shortcut rather than validating travel history and alibis.
The Fargo Police Chief said the department does not operate its own AI facial recognition tool, will avoid using West Fargo’s system, and plans staff training and policy reforms to prevent future misidentifications.
Lipps was extradited on a mid-summer date and spent more than three months in a Tennessee jail before charges were dismissed without prejudice to allow for further investigation, with her release following the dismissal.
The case contributes to national scrutiny of AI in policing, underscoring risks of misidentification and wrongful detention amid rapid AI adoption.
Officials admitted procedural shortcomings, including not submitting relevant surveillance photos to state and local intelligence centers, and pledged reforms such as monthly reviews of facial recognition identifications and closer collaboration with state and federal resources.
Angela Lipps, a Tennessee grandmother, spent more than five months jailed after North Dakota police used an AI tool to tie her to Fargo bank fraud she denies visiting the state for.
West Fargo Police used Clearview AI to identify a suspect but failed to produce sufficient evidence to charge anyone; Fargo police later said they would discontinue using information from West Fargo’s AI system.
Investigators later uncovered exculpatory bank records showing Lipps was in Tennessee during the alleged crimes, and the case was treated as open and under review.
Summary based on 1 source
