The tech world moves so quickly that dystopia can arrive without us realizing it.
That seems to be the trajectory we're on with facial recognition. Some of the biggest names in the tech space are developing advanced facial recognition systems and selling them to governments around the world, but the general public might not fully understand all the risks they bring.
With prominent politicians like New York representative Alexandria Ocasio-Cortez speaking out about the dangers of facial recognition, it's time to get caught up on why so many people are so concerned about the controversial tech.
It's inherently creepy
The first and most obvious reason why people are unhappy about facial recognition is that it's unpleasant by nature. Increasing government surveillance has been a hot-button issue for many, many years, and tech like Amazon's Rekognition software is only making the dystopian future feel even more real.
Rekognition has been used in police trials in the United States as a way to let law enforcement scan faces in public spaces and match them up with those in police databases. It sounds like something a villainous, tyrannical government would do in a science fiction story, but it's happening in real life.
Based on that alone, it's not tough to see why people want to push back against it.
It can happen without your consent
Another big point of contention is that facial recognition tech is being developed and potentially used without the consent of the general populace. This is not new in tech; it's pretty much a given that any data you give to social networks is being used in questionable ways you probably aren't fully aware of, for example.
However, it's a different issue when it comes to your face. Earlier in 2019, we found out IBM had been scraping images from Flickr to train its facial recognition AI without telling people featured in photos on the platform. In addition, U.S. Customs and Border Protection has been collecting photos of people entering and leaving airports for its own facial recognition program.
Some of those photos were taken in a data breach in May, which probably didn't make anyone feel better about their data being collected en masse.
It doesn't even always work properly
Since facial recognition systems are being deployed in real-world scenarios, you would assume they at least work correctly all of the time. That assumption would be wrong, unfortunately.
An ACLU study in 2018 found that Amazon's Rekognition software incorrectly matched members of Congress with other people who appeared in police mugshots. A disproportionate number of those false positive belonged to congresspeople of color, highlighting one of the biggest concerns with facial recognition: It seems pretty biased against people who aren't white men.
Researchers from MIT came to a similar conclusion in 2018, finding that errors in facial recognition shot up significantly when applied to people of color. The general line of thinking is that facial recognition as it exists now reflects the biases of the people who create it.
Considering Rekognition has been deployed by real police departments and has supposedly been marketed to ICE, it stands to reason that its issues with people of color could lead to huge problems with law enforcement. Maybe the technology will get better, and at least one company, Microsoft, has acknowledged the problem, but that's not especially comforting to those who could be hurt by it.
It hasn't been properly regulated
Facial recognition comes with a bunch of problems, as outlined above. That said, it might be a little more palatable to people with comprehensive federal oversight to make sure it isn't abused. Unfortunately, that's not the case right now.
The U.S. government hasn't passed broad regulation for facial recognition yet, as the technology is still new enough that it's a developing issue in the eyes of many. In a series of congressional hearings in May, there was rare bipartisan support for facial recognition legislation, which could be slightly encouraging.
Of course, that probably won't be a perfect solution, even if it comes to pass. Plenty of people likely wouldn't trust the current Congress and presidential administration to craft the best legislation for facial recognition.
Government solutions may not have to come at the federal level, though. Somerville, Mass. and San Francisco have both banned facial recognition in law enforcement at the city level. Local politicians might have to step in and do something about this while the federal government figures out what to do.
Topics Amazon Facial Recognition Privacy
