Police used Flock AI cameras to accuse a Denver woman of stealing a $25 package. It took her weeks to prove she was never there
Approximately 139,000 Flock cameras are up and running in the U.S., and one of them recently turned a Colorado woman’s life upside down.
The cameras, known as automated license plate reading (ALPR) systems built by Atlanta-based Flock Safety, help law enforcement by recording and storing passing vehicle license plates, vehicle manufacturers and models, and any special features in a national database. The data can be used by police departments to track down and arrest criminals for transport-related offenses like kidnapping, bank robberies, and armed robberies.
Chrisanna Elser, a Denver resident, found out the hard way when a police officer rang her doorbell and accused her of stealing an Amazon package.
“You know why I’m here,” Columbine Valley Police Sgt. Jamie Milliman told Elser, who recorded the engagement on her doorbell video device, as reported by CBS News Colorado.
“You know we have cameras in that town. You can’t get a breath of fresh air in or out of that place without us knowing, correct?”
Elser said the conversation started with Milliman looking to confirm if her truck, a green Rivian, was the same vehicle seen driving through the nearby town of Bow Mar that was tied to the theft case. It escalated when Milliman produced photos alleging it was Elser in the vehicle identified by the police and tied her to the theft. “It went right into, ‘we have video of you stealing a package,’” Elser said, noting that Milliman issued a summons before leaving her property.
Flock cameras are gaining notoriety, but for the wrong reasons
Two weeks later, after Elser provided video images from her Rivian, which recorded her whereabouts on the day noted in the summons, the case was dropped by the Columbine Valley police chief.
Privacy experts say they expect more cases like Elser’s, mostly due to the rising use of Flock camera technology, the conclusions investigators draw from the data and the failure to adequately investigate alternative explanations.
“By placing too much reliance upon a ‘product’ and using it as a crutch to rush an investigation without proper operating procedures being followed, we’ll see an increase in these types of cases,” Philip Farina, founder and chief security and safety advisor for Risky Business, a safety and security company, told Moneywise.
Flock camera misuse could also trigger an avalanche of legal claims that can arise from law enforcement getting it wrong. “False arrest or imprisonment, malicious prosecution, civil rights violations are just a few claims that can be made,” Farina said.
Those lawsuits can name the law enforcement officer, their agency, the city/county and the dealer and manufacturer of the Flock cameras. “Damages can include compensation for emotional distress, mental anguish, lost wages and reputational damage,” Farina noted.
Law enforcement’s investigative game
Data security analysts say there’s no indication the Flock technology malfunctioned in the Elsner case. Instead, investigators developed tunnel vision about who could have committed an offense.
“Here they seem to have assumed the car owner committed an offense as her body shape apparently loosely resembled that of another person,” Nora Demleitner, a senior fellow at the Council on Criminal Justice and a long-time criminal justice scholar, told Moneywise. “The danger is that once a theory is adopted, it’ll be made to fit, however uneasily. Questioning basic assumptions is essential throughout a case.”
Another thorny issue with video technology involves administrative ease. “It may feel easier to move the case onto the person accused to prove her innocence,” Demleitner noted. ”Yet, that is not how the law operates, as the officer’s comments were incorrect. One piece of evidence doesn’t make a case.”
More Flock technology training for law enforcement personnel would go a long way toward catching crooks while keeping privacy rights a priority.
“Good ideas include training on the functioning of these systems and their limitations for all officers and prosecutors,” Demleitner said. “Other helpful strategies include independent accountability and oversight through citizen review boards; cities setting out clear rules on what level of supervision they will tolerate, weighing public safety against privacy rights.”
The legal and law enforcement system could also produce an auditable decision trail that can be opened to see what has gone wrong or right. “The public and others should understand that this is not a black box,” Olga V. Mack, CEO at TermScout, a legal system contract services firm, told Moneywise. “They need to understand how the alerts are triggered. What was it? What was the data source? How confident are we that the information is available, reliable, and hasn’t been tampered with? Who reviewed it? Are they trained?”
Also, there needs to be some form of due process where people not only understand what triggered an accusation and who reviewed it, but also how they can challenge it. “That would include knowing how Flock users in law enforcement, before and after suspicion occurs, present evidence,” Mack noted.
–
Original Article: Police used Flock AI cameras to accuse a Denver woman of stealing a $25 package. It took her weeks to prove she was never there
About Post Author
