Police departments across the nation are generating leads and making arrests by feeding celebrity photos, CGI renderings, and manipulated images into facial recognition software.
Often unbeknownst to the public, law enforcement is identifying suspects based on “all manner of ‘probe photos,’ photos of unknown individuals submitted for search against a police or driver license database,” a study published on Thursday by the Georgetown Law Center on Privacy and Technology reported.
The new research comes on the heels of a landmark privacy vote on Tuesday in San Francisco, which is now the first US city to ban the use of facial recognition technology by police and government agencies. A recent groundswell of opposition has led to the passage of legislation that aims to protect marginalized communities from spy technology.
These systems “threaten to fundamentally change the nature of our public spaces,” said Clare Garvie, author of the study and senior associate at the Georgetown Law Center on Privacy and Technology.
“In the absence of rules requiring agencies to publish when, how, and why they’re using this technology, we really have no sense [of how often law enforcement is relying on it],” Garvie said.
The study provides several real-life examples of police using probe photos in facial recognition searches.
On April 28, 2017, a suspect was caught on camera allegedly stealing beer from a CVS in New York City. When the pixelated surveillance footage produced zero hits on the New York Police Department’s (NYPD) facial recognition system, a detective swapped the image for one of actor Woody Harrelson, claiming the suspect resembled the movie star, in order to generate hits.
Harrelson’s photo returned a list of candidates, including a person who was identified by NYPD as the suspect and arrested for petit larceny.
The NYPD also fed an image of a New York Knicks player (whose name was redacted in police documents provided to the Georgetown Law Center on Privacy and Technology) into its facial recognition system in order to find a Brooklyn man wanted for assault.
These are just two cases of police trying to brute force leads from facial recognition technology using dubious images.
NYPD records show that it made 2,878 arrests stemming from facial recognition searches in five years. Florida law enforcement agencies run roughly 8,000 of these searches per month, according to documents from the Pinellas County Sheriff’s Office.
Departments are also using poorly-drawn probe images in facial recognition searches, such as a forensic sketch that was fed into Amazon’s facial recognition software, Rekognition, by police in Washington County, Oregon…….Read more>>