Detroit Police Wrongfully Arrested Black Man Due To Faulty Facial Recognition

Detroit Police Department used facial recognition technology developed by DataWorks Plus to identify suspects in criminal investigations starting in 2017. In May 2019, police used the system to analyze a cellphone video of a larceny incident and it incorrectly identified Michael Oliver, a 25-year-old Black man with visible tattoos, as the suspect despite the actual perpetrator having no visible tattoos and different physical characteristics. Oliver was arrested in July 2019, charged with felony larceny, and held for 2.5 days before the case was dismissed when evidence clearly showed he was not the perpetrator. A similar case involved Robert Williams, a 42-year-old Black man wrongfully arrested in January 2020 for allegedly stealing watches. Detroit Police Chief James Craig acknowledged that the facial recognition technology misidentifies suspects 96% of the time when used alone. Studies show the technology has substantially higher error rates for people of color, and Detroit's population is nearly 80% Black. The Wayne County Prosecutor's Office has since implemented stricter protocols requiring supervisor review of all facial recognition cases before charging decisions. Oliver filed a lawsuit seeking at least $12 million in damages, citing loss of employment, housing, and other life disruptions caused by his wrongful arrest.