I-17 101 traffic interchange

Arrest shows how facial recognition helps ADOT prevent identity theft

Arrest shows how facial recognition helps ADOT prevent identity theft

April 5, 2017

PHOENIX – A Casa Grande resident who allegedly used stolen identities to fraudulently obtain benefits and settlement payments was arrested thanks to Arizona Department of Transportation detectives’ use of facial recognition training and technology.

The case began when Sharon Forrest, 51, applied for a driver license in early February. A review found that her photo closely resembled two other photos in ADOT’s database, and detectives with FBI training in facial recognition determined that all three photos were of Forrest.

Their investigation found that Forrest had used stolen identities to get Social Security benefits, Department of Economic Security assistance, settlement payments and veteran benefits. She was arrested March 1 and booked at the Maricopa County Sheriff’s Office Fourth Avenue Jail on eight counts of forgery and one count of identity theft.

This case is just one example of how facial recognition technology used by ADOT’s Office of Inspector General protects Arizonans’ identities and helps prevent fraud involving state-issued driver licenses and identification cards.

“As the agency that issues identification cards, we’re serious about protecting Arizonans’ identities,” said Michael Lockhart, chief of the Office of Inspector General. “This system helps us to do just that.”

Facial recognition allows detectives to compare a photo against others in the driver license database to ensure a person isn’t fraudulently obtaining an ID card.

If a photo is a likely match to another one, the system will flag it. Potentially fraudulent photos then undergo three levels of review by detectives who have received FBI facial recognition training.

“Humans are involved at every level of facial recognition technology,” Lockhart said. “We make sure we’ve got solid evidence before advancing one of our cases. This is not a science fiction scenario where the computer determines potential fraud all by itself.”