by Emily Tran
What is the line between using technology to make our society secure without living in a security state? Have we already crossed it with facial recognition?
San Francisco, one of the most tech-friendly and tech-savvy cities in the world, has publicly announced that it believes we have crossed it. This is now the first city in the United States to ban the use of facial-recognition technology by the city’s 53 departments—including the San Francisco Police Department. However, exceptions still hold for airports, businesses, and residents that are permitted to use general facial recognition or surveillance technology such as security cameras.
Aaron Peskin, a city supervisor and the sponsor of the bill, is concerned that technology has become fundamentally invasive and psychologically unhealthy when people know they are being watched in every aspect of the public realm.
While there have been major improvements to facial recognition, there is a significant portion of people that share similar opinions to Peskin. Is this a large enough concern to prohibit the use of it entirely? In this blog, we will explore how facial recognition has grown and whether or not society needs to pull in the reins on the power of technology.
What Exactly Is Facial Recognition?
Facial recognition is a biometric software application that uses a person’s digital image to verify that a legitimate user has access to confidential or sensitive personal information. It is a type of augmented identification to who you actually are instead of things you memorize (passwords) or carry (driver’s license).
Facial recognition functions by measuring nodal points, the distances between major structural pieces on your face to make a faceprint. A faceprint is a digitally recorded representation of a person’s face that acts as a unique code to then help technology connect faces to names. With the advances in skin texture analysis, it can now measure the distance between exact pores which can be so precise as to tell the difference between twins.
Facial Recognition Has Taken Off
According to MarketWatch, the global facial recognition market had been valued at $3.04 billion in 2016 and is expected to reach $8.93 billion by 2022. But where do we see this form of technology being used? The public’s increasing exposure to facial recognition can be largely attributed to Apple’s FaceID and filters on social media. This was likely the tipping point for the public acceptance of facial biometrics. However, unlocking your phone and distorting features are only inklings of its scope and the possibilities that facial recognition has uncovered.
The value of facial recognition spans far into both digital and physical security. Digitally, it has provided a platform for fraud detection with facial liveness features and frictionless authentication with AI. Physically, these features have heightened threat and intrusion detection, perimeter and asset monitoring, and known individual detection.
Security forces with this technology built-in can check identities against their databases in seconds. Online, it has positively helped verify and authenticate people to mitigate against the current cybercrime crisis. Buildings have used this as a more reliable option, whereas access cards, keys, and badges can be borrowed, copied, or stolen. Banks have viewed this as an opportunity for fraud prevention and safeguarding vaults. According to the New York Times, internal police departments have documented over 8,000 cases involving facial recognition as an investigative tool in 2018 alone. Even airport security implemented variations of facial recognition. A rising successful example is a company named Clear. Clear is a biometric scanning technology used to let travelers skip through airport identification procedures by using a traveler’s fingerprints and eye scans to confirm identity. This company has accumulated over a million users in less than a year.
Although facial recognition is providing better security, it should not be misunderstood as a replacement for other methods of identification but rather a second layer of defense.
Outside of security, countries all over the world have continued to develop methods to integrate facial recognition including, but not limited to, approving age for sales of alcohol and cigarettes, payment methods, tracking attendance in schools, and employee engagement. Businesses have discovered that image and video analytics are strong drivers for this market, and are starting to leverage facial recognition to improve customer service.
The next phase of facial recognition AI is likely to provide emotional analysis on faces to offer businesses a more complete guide to how consumers are interacting with their company and the world around them. This will help enhance marketing strategies and target specific markets.
But even with all these great advancements, facial recognition can still be fooled.
Facing Up to Slip-Ups
Some of the greatest tech giants—Apple, Amazon, and Samsung—have all had their own inaccuracies and blunders when using facial recognition. We have relied on technology to protect and secure our information, but people are actively discovering new ways to trick facial recognition by obscuring the face with hats, scarves, and digital filters. This has planted a new fear of whether or not these big organizations that each claims to only produce the best products understand their own technology in regards to promising privacy. Facial recognition is being used everywhere, but the average citizen has no means to track their faceprint.
A large complication with facial recognition AI across all industries is its relatively high rates of false positives and the consistent cases of racial and gender biases. These mistakes disproportionately affect ethnic minorities that have been falsely identified. Yet, there are very few laws that regulate facial recognition and surveillance. As its dominance continues to grow, will adopting this innovative technology be an infringement on civil liberties? Has security and functionality allowed us to overlook these concerns?
We can look to China for an extreme application of facial recognition, whereby citizens are being assigned a social credit score based on economic and social reputation. Facial recognition AI determines these scores by identifying those that need to be socially punished—throttling your internet speeds, banning you or your kids from the best schools, keeping you out of the best hotels, or putting you on the country’s “blacklist”. Scores can be negatively affected by bad driving, jaywalking, posting fake news online, or buying too many videos games.
Security misidentifications and cases like this one might suggest technology can do more harm than good. It is important that people continue this discussion and remain involved in the decision-making process, and remember that technology has been created to augment the human mind—not replace it.
Bottom Line
While facial recognition AI raises fears of a dystopian surveillance state with the idea of vanishing privacy, it would be an act of ignorance to deny the value of this technology. San Francisco’s concern was valid but to completely prohibit the usage of facial recognition will only position them at a disadvantage as this global market continues to expand. The facial recognition market has been rapidly moving forward and we are predicting unparalleled levels of real-world protection digitally and physically. The next question is to ask whether or not the promise of facial biometrics will deliver and when?