← the record
AIAAIC-0420

Robert Williams wrongly arrested by police using facial recognition

Detroit-based Robert Williams was misidentified by facial recognition technology used by the Detroit Police Department for a crime he did not commit, inflicting significant emotional , psych ological and other damage on Williams and his family. What happened Williams, 42, had been arrested for reputedly stealing five high-end watches at a store in Detroit . A detective used facial recognition technology on a grainy image from the store's CCTV video, and the system flagged Williams as a potential match based on a driver’s license photograph. A security guard who had not been present at the incident then identified Williams in a photo line-up of Black males. Arrested in front of his family, Williams was arrested, arraigned, detained for 30 hours and questioned in connection with a crime that took place in a store he had not visited since 2014. The incident resulted in Williams experiencing severe emotional trauma, loss of dignity and social s tigmatisation from his family and community. In addition, he suffered racial discrimination, a loss of right to liberty and security, and loss of privacy. He also had to cover his legal fees, resulting in financial/earnings loss. Why it happened William's wrongful arrest highlighted well-known issues with facial recognition technology, notably its high error rates and biases against minority groups, with studies showing that these systems often misidentify individuals with darker skin tones, leading to disproportionate impacts on Black and other marginalised communities. In this case, the Detroit Police Department's reliance on this technology without sufficient corroborating evidence contributed to the wrongful arrest. Detroit Police Chief James Craig had earlier admitted facial recognition technology misidentified suspects in 96 percent of cases. What it means Robert Williams' wrongful arrest case resulted in a landmark legal settlement with the Detroit Police Department, which was compelled to commit to reforms regarding its use of facial recognition technology. The case also sparked discussions about the ethical implications of using facial recognition technology in law enforcement, and raised questions about accountability and the need for stricter regulations governing the use of such technologies to prevent future injustices. The incident prompted calls for reform in policing practices and greater scrutiny of technological tools used in criminal investigations. System 🤖 DataWorks Plus Developer: DataWorks Plus Country: USA Sector: Govt - police Purpose: Identify criminal suspect Technology: Facial recognition Issue: Accountability; Accuracy/reliability; Human/civil rights; Privac y/surveillance ; Transparency Timeline ⏰ October 2018 . Theft of watches occurs at a Shinola store in Detroit, captured on surveillance video. March 2019. Michigan State Police examiner Jennifer Coulson uploads a still image from the video into the state's facial recognition database, generating a possible match to Williams's expired driver's licence photo. January 9, 2020. Williams is arrested at his Farmington Hills home in front of his wife and two children and held for roughly 30 hours. July 1, 2020 : US Senator Sherrod Brown sends a letter to DataWorks Plus demanding answers about its technology's use nationwide. August 2020. The New York Times publicly reports the wrongful arrest. April 2021. Williams sues the Detroit Police Department for wrongfully arresting and jailing him. He was the first person known to have been arrested in the US because of a facial recognition failure. June 2024. The City of Detroit settles with the ACLU and Robert Williams for USD 300,000. Legal, regulatory👩🏼‍⚖️ Williams v City of Detroit

Date it happened
2018-10-01
Organisation involved
Detroit Police Department
Product, system or model
DataWorks Plus FR
Where this came from
Share this incident
XLinkedInFacebookWhatsAppEmail
Attribution

This incident was imported from AIAAIC and is used under CC BY-SA 4.0. Our additions to it — the structured fields, the translation, the checks against other reports — are published under the same licence.

This is a record of what was reported, not a finding that anyone broke the law. If it names your organisation and you believe it is wrong, the corrections process is free and open to everyone.