The city of Detroit has agreed to pay $300,000 to a man who was wrongly accused of shoplifting due to a flawed facial recognition technology match. The settlement also includes changes to how the city's police force uses facial recognition technology to identify suspects.
The man, Robert Williams, had his driver's license picture incorrectly flagged as a likely match for a suspect captured on security video at a watch store theft in 2018. Williams, a Black man, was arrested in front of his family two years later, only to have the charges eventually dropped after spending more than 24 hours in jail.
Following the incident, Williams expressed his hope for a better world with improved safeguards on the use of facial recognition technology. The American Civil Liberties Union (ACLU) highlighted the flawed and racially biased nature of the technology, citing a higher rate of false matches for Black individuals.
As part of the settlement, Detroit police are now prohibited from making arrests based solely on facial recognition results and photo lineups created from such searches. Police Chief James White introduced new policies on facial recognition technology last August, emphasizing the need for additional evidence beyond technology to establish a suspect's involvement in a crime.
The settlement also includes an audit of all cases from 2017 to 2023 that relied on facial recognition technology to obtain arrest warrants. The ACLU praised the agreement, stating that it should prevent future wrongful arrests and serve as a model for other police departments using facial recognition technology.
Phil Mayor, a senior staff attorney at the ACLU of Michigan, emphasized the importance of implementing guardrails to limit the misuse of facial recognition technology by law enforcement. The ACLU represented Williams, who was also supported by the Civil Rights Litigation Initiative at the University of Michigan Law School.
Michael J. Steinberg, director of the Civil Rights Litigation Initiative, expressed hope that the settlement would prevent further wrongful arrests and commended Williams for his role in advocating against the misuse of facial recognition technology.
