A facial recognition glitch leads to the wrong man being arrested and jailed for SIX days – despite never visiting Louisiana, where a purse robbery was committed and weighs 40 pounds less than a real criminal
- Randall Reid, 28, was wrongly jailed in November after facial recognition technology mistook him for a handbag theft in Louisiana
- Aside from a facial mole and a 40-pound weight difference from the suspect, Reid says he’s never been to Louisiana
- Reid is black and critics warn that facial recognition technology may lead to higher rates of misidentification for people of color
Louisiana authorities’ use of facial recognition technology led to the erroneous arrest of a Georgia man on a fugitive warrant, said a lawyer in a case drawing attention to racial disparities in the use of the digital tool.
Randall Reid, 28, was jailed in DeKalb County, Georgia on Nov. 25 after authorities misidentified him as a pickpocket in Jefferson Parish and Baton Rouge.
“They told me I had a search warrant from Jefferson Parish. I said, “What’s Jefferson Parish?” Reid said. “I’ve never been to Louisiana a day in my life. Then they told me it was for theft. So not only was I not in Louisiana, I don’t steal.’
He was released on December 1 when authorities realized their mistake.
Reid is black, and his arrest draws new attention to the use of a technology that critics say leads to higher misidentification rates for blacks than for whites, The Times-Picayune/The New Orleans Advocate reported.
Critics warn that facial recognition technology may lead to higher rates of misidentification of people of color. Pictured: a random man
Tommy Calogero, Reid’s attorney, said he was falsely linked to the theft of luxury purses from a consignment store in Metairie, a New Orleans suburb in Jefferson Parish.
The thieves stole $10,000 worth of luxury Chanel and Louis Vuitton bags in three days.
A Baton Rouge Police Department detective then assumed the Jefferson Parish Sheriff’s Office’s identification of Reid in order to obtain an arrest warrant claiming he was among three men involved in another luxury purse theft that same week, court records show show according to the newspaper.
Differences, like a mole on Reid’s face, prompted the Jefferson sheriff to cancel the warrant, said Calogero, who estimated a 40-pound difference between Reid and the pickpocket in surveillance footage.
“I think they realized they got on their feet and made an arrest based on a face,” Calogero told NOLA.
The police’s use of facial recognition in Reid’s arrest is unclear.
Tommy Calogero, Reid’s attorney, said he was wrongly linked to the theft of luxury purses from a consignment shop in Metairie in June
Reid’s case draws renewed attention to the use of facial recognition tools in Louisiana and elsewhere.
In July, the New Orleans City Council voted to allow police to use facial recognition after several people complained about privacy issues, NOLA reported.
Police can use facial recognition to identify violent crime suspects after all other tactics have failed.
Authorities in New Orleans say facial recognition can only be used to generate leads and officials must obtain approval from department officials before submitting an application through the Louisiana State Analytic and Fusion Exchange in Baton Rouge.
Under the latest city rules, all possible matches must be peer-reviewed by other facial recognition investigators.
Legislation restricting the use of facial recognition across the country died in a 2021 legislature.
HOW DOES FACE RECOGNITION TECHNOLOGY WORK?
Face recognition software works by matching real-time images to a previous photograph of a person.
Each face has approximately 80 unique nodes across the eyes, nose, cheeks, and mouth that distinguish one person from another.
A digital video camera measures the distance between various points on the human face, such as the width of the nose, the depth of the eye sockets, the distance between the eyes, and the shape of the jaw.
This creates a unique numeric code that can then be linked to a matching code from a previous photo.
Facial recognition systems have been criticized for their mass surveillance capabilities, which raise privacy concerns, and because some studies have shown that the technology is far more likely to misidentify blacks and other people of color than whites, leading to false arrests.
The research takes place amid the widespread deployment of facial recognition technology for law enforcement, airports, banks, retail and smartphones.
Mistakes could result in arresting the “wrong people” and conducting “lengthy interrogations,” according to Jay Stanley of the American Civil Liberties Union.
A 2019 study by the National Institute of Standards and Technology (NIST) found that two algorithms misassigned black women 35 percent of the time.
Activists and researchers have claimed the potential for error is too great and mistakes could lead to the imprisonment of innocent people.
They also claimed that the technology could be used to create databases that could be hacked or used inappropriately.
The NIST study found both “false positives,” where an individual is incorrectly identified, and “false negatives,” where the algorithm does not accurately match a face to a specific person in a database.
An MIT Media Lab facial recognition software expert says this study shows that the spread of facial surveillance should be stopped to protect people.