Hays County approves AI facial recognition technology for sheriff’s office
HAYS COUNTY, Texas (KXAN) — A facial recognition software could come to the Hays County Sheriff’s Office in the future.
County commissioners recently approved for the sheriff’s office to fill out an order form with Clearview AI, a facial recognition company that provides software to law enforcement, government agencies and other organizations.
The agenda item said the sheriff’s office is approved to purchase Clearview AI software licenses in the FY 2024 budget.
“The software is a facial recognition software that will provide law enforcement with greater insight and lead generation through the use of its investigative platform,” the agenda item said.
It said the term for this purchase is over a three-year period, with $6,495 due each year. It totals out to $19,485.
The agenda item said funding for the first year is available in the Sheriff’s Office Software Capital general ledger.
KXAN reached out but the sheriff’s office was not available for comment.
What is Clearview AI?
According to Clearview AI’s website, its platform has the “largest known database of 40+ billion facial images sourced from public-only web sources, including news media, mugshot websites, public social media, and many other open sources.”
When it comes to law enforcement, the company said the platform helps with developing leads.
The technology allows law enforcement agencies to feed images from video surveillance into the software that can search for a possible match.
It claimed on its website that the leads, along with other evidence, can help identify suspects, persons of interest and victims to help solve crimes.
The CEO of Clearview AI sent a statement to KXAN about its uses.
“AI is used by law enforcement agencies around the country for after-the-crime investigations to help identify criminal suspects and help solve crimes such as drug trafficking, human trafficking and child abuse. Clearview AI only searches publicly available information from the internet. It is not intended to be used as a real-time surveillance tool. Clearview AI requires its law enforcement customers to provide a case number and crime type to ensure an audit trail and to enforce responsible usage of facial recognition and provides training for its customers.”
Hoan Ton-That, CEO of Clearview AI
Misidentification of people of color
Dr. Peter Stone is the founding member of Good Systems, a team of researchers at the University of Texas at Austin focused on ethical AI systems.
PREVIOUS COVERAGE: New UT program puts focus on ethical AI, building robots to do ‘helpful tasks’
“Many AI technologies, with facial recognition, in particular, have the potential to be beneficial and to lead to better law enforcement, but they also have the potential for misuse,” Stone said.
Stone said this technology comes with risks.
“They’re not always equally capable with all demographics, with all skin colors. There’s potential to exacerbate biases, to be misused,” Stone said. “There have been well publicized examples of both bias and wrongful arrests based on facial recognition determinations.”
Dr. Samantha Shorey works alongside Stone with Good Systems.
She mentioned a project called Gender Shades by researcher Joy Buolamwini that studied the way different facial recognition tools identify people.
“She found that facial recognition is a lot better at identifying males than females and a lot better at identifying people with lighter skin that people with darker skin,” Shorey said. “These classification algorithms performed the worst around darker skinned females.”
Shorey said this can result in a higher rate of misidentification of people of color.
Wrongful arrest and privacy concerns
Top lawmakers are also calling this technology into question.
U.S. Senator Ed Markey (D-Massachusetts) sent a letter to Clearview AI asking questions about the company’s technology.
According to a press release from the senator, he expressed “serious concerns about Clearview AI’s mass collection of biometric information without individual consent, arguing that Clearview AI’s practices pose serious threats to privacy rights and civil liberties, particularly for people of color.”
Facial recognition technology jailed a man for days. His lawsuit joins others from Black plaintiffs
The senator also referenced a report about a Louisiana Sheriff’s Office relying on incorrect Clearview AI results in the wrongful arrest of a Black Georgia resident who said he never visited Louisiana.
Ultimately, Shorey and Stone said law enforcement should use it with caution.
“Be familiar with what the limitations of the technologies are, what the risks of the technologies are,” Stone said. “And to certainly not use it in a way that takes away or supplants human judgment.”
Oversight of the technology
When it comes to ownership over this tool, Shorey said it’s going to “consolidate power in the hands of the companies that are selling it, and the agencies that can afford to use it.”
She said there’s no democratic management over facial recognition technology.
Stone suggested some sort of oversight to make sure it isn’t being misused.
“Having a third party validator or some kind of independent entity that makes sure that what they’re saying is actually being put into practice,” Stone said.
For the latest news, weather, sports, and streaming video, head to KXAN Austin.
