Porcha Woodruff was eight months pregnant when she was falsely arrested for a carjacking. The evidence that led to the arrest of the Detroit mother of three was facial recognition technology, which was flawed. She is the first woman in the United States to report being falsely arrested on evidence using faulty facial recognition.

Ms. Woodruff was having a quiet morning at home with her children when six police officers came to her door and told her she was under arrest for robbery and carjacking. The pregnant mom said that she thought at first it must be a joke.

When they informed her about the arrest, she said to the police, “Are you kidding, carjacking? Do you see that I am eight months pregnant?”

At the time of her arrest, Ms. Woodruff was getting her two children ready for school. When the police officers arrested her, she turned to them and said, “Mommy is going to jail.”

Despite the fact her fiancé rushed to her defense while she was being arrested, she was taken away to police headquarters to be booked.

The shocked mother, whose upset six and twelve-year-olds were crying at the door upon her arrest, filed a civil complaint with the Southern Division of the Eastern District of Michigan.

The heavily pregnant woman filed a complaint against the city of Detroit and Oliver. In her suit, Ms. Woodruff alleges that “despite knowing” that Woodruff “was not involved in the robbery or carjacking, Detective Oliver sent her to return to the holding cell, after being arrested.

She was held for eleven hours, in grim conditions for a pregnant woman, after her arrest. Court documents say that after waiting for approximately two hours following her interview with Detective Oliver, Ms.Woodruff was arraigned, and charged with robbery and hijacking. She was then released on a $100,000 personal bond at about 7 p.m.

Ms. Woodruff, despite being eight months pregnant, was held for eleven hours in detention. The police kept her sitting on a hard, concrete bench because they claimed that there were no beds or chairs in the detention center. In her civil complaint and medical records, the heavily pregnant, innocent woman experienced severe cramping she described as “whole belly tightening” every ten to fifteen minutes.

“I was having contractions in the holding cell. My back was sending me sharp pains. I was having spasms,” said Ms. Woodruff.

Ms. Woodruff’s fiancé drove her immediately to the hospital upon her release. Doctors found that her heart rate had dipped and she was dehydrated.

Though the AI facial recognition identified Ms. Woodruff as the carjacker, it was wrong. Ms. Woodruff slightly resembles another Black woman who allegedly did the robbery and carjacking. That suspect, who did not appear to be pregnant, did not show up at her court date and has not been found.

The faulty facial recognition was created at a BP gas station and supposedly matched an old photograph of Ms. Woodruff taken when she was charged with driving with an expired license, a misdemeanor. Carjacking, however, is a felony.

The facial recognition AI did not use the newer, accurate photo of Ms. Woodruff from her 2021 license, but instead, used the photo from eight years prior, when she was in her twenties.

To make matters worse, the victim in the carjacking, who may have been drugged by the culprit, identified Ms. Woodruff from six photos shared by the police.

In the United States, six other people have been misidentified by AI facial technology. They are all Black. The ACLU reports that fifty percent of the falsely accused black people are from Detroit and have been arrested by Detroit Police.

The other falsely arrested people accused by police of crimes based on faulty AI facial recognition are participating in a separate lawsuit. In particular, court documents alleging that falsely arrested Robert Williams was the victim of wrongful arrest and imprisonment state that his “case exemplifies the grave harm caused by the misuse of, and reliance upon, facial recognition technology. Plaintiff Robert Williams was falsely arrested because, as Detroit police officers later admitted, “the computer got it wrong” and erroneously identified him as the suspect in a watch theft investigation. Nonetheless, officers secured a warrant for Mr. Williams’s arrest without providing the authorizing magistrate with critical information about deficiencies in the investigation and how facial recognition technology was used.”

Facial recognition can be erroneous and lead to false arrests or imprisonment. Nathaniel Erb, state policy advocate at the Innocent Project, says that facial recognition “increases the risk that innocent people will be wrongfully convicted.” Founded in 1992 by visionary attorneys Peter Neufeld and Barry Scheck, the Innocence Project has been described as being at the forefront of criminal justice reform, using DNA and other scientific advancements to prove wrongful convictions.

The Innocent Project is pushing for all AI facial technology to be put on pause until the technology can be perfected.

Facial recognition technology has been found to share false information when used to identify Black people. The White House Office of Science and Technology Policy reported in the Blueprint for an AI Bill of Rights that this AI technology can falsely target Black people. Another study conducted by the National Institute of Standards and Technology reports that Black and Asian people are 10 to 100 times more likely to be “misidentified” by such technology than White people.

Beyond racial bias created by faulty facial recognition, the Innocent Project reports that “these technologies often store or exploit personal data, leaving vulnerable communities, especially those that have been historically criminalized, exposed to data harms.” The report continues, “More often than not, these tools also enable tunnel vision among law enforcement, encouraging investigators to hone in on a specific individual or individuals as people of interest even in the presence of convincing exculpatory evidence.”

A month after her wrongful arrest, all charges were dropped against Ms. Woodruff. In spite of the ordeal, she gave birth to a healthy son.