Amazon, Google parent Alphabet and Microsoft used people's photos to train their facial recognition technologies without obtaining the subjects' permission, in violation of an Illinois biometric privacy statute, a trio of federal lawsuits filed Tuesday allege.
The photos in question were part of IBM's Diversity in Faces database, which is designed to advance the study of fairness and accuracy in facial recognition by looking at more than just skin tone, age and gender. The data includes 1 million images of human faces, annotated with tags such as face symmetry, nose length and forehead height.
The two Illinois residents who brought the lawsuits, Steven Vance and Tim Janecyk, say their images were included in that data set without their permission, despite clearly identifying themselves as residents of Illinois. Collection, storage and use of biometric information is illegal in the state without written consent under the Biometric Information Privacy Act, passed by the Illinois legislature in 2008.
The defendants "chose to use and profit from biometric identifiers and information scanned from photographs that were uploaded from Illinois; managed via Illinois-based user accounts, computers and mobile devices, and/or created in Illinois," the lawsuits say. "In doing so, [defendants] exposed Illinois residents and citizens to ongoing privacy risks within Illinois, knowing that [their] conduct would injure those residents and citizens within Illinois."
Facial recognition has faced backlash from privacy advocates and lawmakers, and a handful of cities have banned the municipal use of the technology. Last year, Democratic lawmakers proposed prohibiting public housing units from using technology like facial recognition. Still, facial recognition systems are on track to become pervasive in airports and shopping centers, and some companies are selling them to police departments.
Critics cite studies showing that the technology has low accuracy rates for women and minorities. When it works properly, they add, it has the potential to become an inescapable and invasive form of surveillance. Companies such as Clearview AI, which has technology that lets users identify people by comparing their faces to photos scraped from the internet, have raised concerns over the power of the technology.
IBM, which collected the images from the photo site Flickr, faced criticism earlier this year from some photographers, experts and activists for not informing people their images were being used to improve facial recognition technology. In response, IBM said it takes privacy seriously and that users could opt out of the data set.
IBM has been a strong proponent of regulating facial recognition, but in June the company announced it's, saying it worries that the technology is being used to promote discrimination and racial injustice.
The lawsuits, filed in California and Washington state courts where the companies are based, seek class-action status, as well as monetary damages and restraint of defendants' activities pertaining to the database.
Microsoft said it had received the complaint and is reviewing it.
"We take privacy seriously and we are committed to ensuring our AI technology is developed and used responsibly," a Microsoft spokesperson said.
Representatives for Amazon and Google didn't respond to requests for comment.