Facial Recognition and the Opportunities for Better Customer Experience
In George Orwell’s epic novel 1984, our protagonist Winston Smith is constantly aware that he must toe the government’s line because “Big Brother is watching.” When Orwell’s dystopian work was first published in the middle of last century, the technology needed to monitor citizens’ coming and going was in its infancy; but now, some seven decades later, it is becoming mainstream.
A major cog in that technology is facial recognition (FR). Backed by the proliferation of high-resolution cameras in public spaces, robust artificial intelligence algorithms, and ever-increasing computer processing capabilities, FR has quickly become entrenched in our everyday lives.
Many privacy advocates are alarmed. “Body-Worn Cameras should NEVER have facial recognition capabilities,” tweets Privacy International. “This would subvert the purpose of body-worn cameras as a tool of police accountability & transparency by turning them into a tool of mass surveillance.” Likewise, several US congressional representatives have called for restrictions or even moratoriums on the use of FR by law enforcement. “Seems to me it’s time for a timeout,” says US Rep. Jim Jordan (R-Ohio), citing First and Fourth Amendment concerns. “Doesn’t matter what side of the political spectrum you’re on, this should concern us all.” The cities of San Francisco and Somerville, MA have banned local police and city agencies from using the technology; others are likely close behind.
Furthermore, skeptics cite concerns related to inaccuracy and potential biases. Some studies have indicated that FR delivers more accurate results on male Caucasians than on females and darker-skinned individuals. The jury is out as to whether results are skewed due to the data used to train the models, the makeup of the FR workforce, or other factors. One thing that virtually all experts agree on is that the technology is relatively immature and will continue to advance. As with other areas of artificial intelligence, there is a significant focus on proactively working to ensure that FR avoids magnifying existing human biases.
But while concerns related to privacy, accuracy, and fairness must be taken seriously, it is also important to consider the benefits we can all realize from applications of FR. An obvious opportunity is security. FR is already used to spot potential threats at major public gatherings such as sporting events. FR is being tested at many major airports and will be deployed at London Heathrow later this year with the promise of both improving security and reducing reliance on ID and boarding passes while cutting boarding times in half. The International Air Transport Association (IATA) says that 64% of passengers would choose to share their biometric data in exchange for a better travel experience.
One of the most promising applications may be in contact centers; more than 70% of incoming calls today originate from mobile phones, not to mention the increasing volume of contacts via SMS, Rich Communication Services (RCS) and messaging channels such as Facebook Messenger and WeChat. Smartphone-based FR can identify and authenticate customers, reducing the need to remember account numbers, user names and passwords.
FR offers a wealth of additional possibilities. It is used today to track attendance at schools and workplaces, to identify photos that we and our friends are in, to unlock phones, authorize payment transactions, and even diagnose diseases. Before long, FR could even replace ATM cards and car keys.
Perhaps the most important takeaway is that – despite the fears of a modern-day Big Brother – FR is not inherently intrusive. As with many technologies, it can be misapplied or even abused; but it can also be applied responsibly to make our lives safer and more convenient. It is up to all of us to let our voices be heard, so that we can guide our policymakers together with the public and private organizations with whom we interact to decide when, where, and how FR will be used to our benefit.