This week Facebook agreed to pay $550 million to settle a class-action lawsuit over its use of facial recognition technology in Illinois. Apparently, Facebook’s photo-labeling service, Tag Suggestions, which uses face-matching software to suggest the names of people in users’ photos, violated an Illinois biometric privacy law.
Despite saying the allegations “have no merit,” Facebook forked over the fine -- pocket change to a company that just announced a 25% increase in revenue to $21.08 billion in a single quarter.
While Facebook collects mug shots of its 2.41 billion monthly active users around the globe, a December report revealed that most commercial facial-recognition systems exhibit bias, falsely identifying African-American and Asian faces 10 times to 100 times more than Caucasian faces.
The tech also has more difficulty identifying women versus men, and falsely identified older adults up to 10 times more than middle-aged adults. So, if you are an older black woman, I’d say your chances of remaining anonymous are pretty good.
This comes at a time when more and more police departments are using facial recognition to try and identify criminals. And AI-based programs are constructing deep-fake faces that don’t exist in the real world, but look every bit as if they do.
All this sounds remarkably like the privacy/accuracy conundrum surrounding digital advertising. Everyone is supposed to be upset that their personally identifiable data is being used to help advertisers serve them more appropriate ads, but if you ever looked under the hood, you would see that a vast majority of tracking data (and its assumptions about you based on algorithms) are mostly wrong.
In fact, the more finite the “audience segment” someone tried to squeeze you into, the more likely the assumptions that got you there are dubious.
So the question comes down to this: If it’s inevitable that your face will be recorded by either online tech or surveillance cameras that are popping up like weeds in a vacant lot, is it in your better interest to have your face removed from databases, or make sure that your mug is correctly labeled as you? You know, the reverse of the police procedurals, where instead of tapping the mugs shots and saying "That’s the guy," you instead tap it and say, "That’s me."
After all, being falsely IDed as a criminal can have life-changing consequences, as noted in a post on The Intercept.
There are privacy advocates who argue that having databases of everyone’s face stored here and there is dangerous, and can unroll some scenarios that are truly frightening should the data fall into the wrong hands.
I know a fair number of people who don’t care that their picture has been collected, but would want to make sure it is a flattering one (NOT that student ID shot from college or their driver’s license).
By the thousands, folks are uploading their most provocative pictures to dating apps -- but somehow don’t expect them to turn up elsewhere on the internet. Fat chance. Angry ex-boyfriends and -girlfriends have proven repeatedly that they cannot be trusted to not share those intimate pictures you sent them when times were better.
One hopes at least those pictures in the databases are only from the shoulders up.