Were you unable to attend Transform 2022? Check out all of the summit sessions in our on-demand library now! Watch here.
Real-time facial recognition systems are no longer speculative – they have become a real-world security solution with several practical, powerful applications. However, the rise of this technology also poses some interesting ethical questions. So, should you be worried about how it will impact daily life?
We’ve all seen, and probably scoffed at, thrillers in which intelligence agencies or the police can zoom in on a face in a blurry security video, make the image crystal-clear with the touch of a button and match that face to one of the 7.8 billion people on Earth in seconds. Facebook users might be familiar with the social network’s own facial recognition software, which performs similar searches on users’ photos, and which has recently been shut down “as part of a company-wide move to limit the use of facial recognition in our products.”
Police departments around the world, and in the U.S. in particular, have been purchasing facial recognition solutions despite their flaws [subscription required]. It would seem that what was previously science fiction is not only reality, but becoming an integral part of modern security systems around the world.
However, advancements in the field raise a number of pressing ethical questions, which the sector must be ready to answer as adoption rates rise. In particular, there is growing concern about the potential for these systems to be used by bad actors to commit fraudulent activity. Privacy campaigners also cite worries about the technology, with others wondering if it may be used to impinge on human rights in the near future. So, the question is whether the rest of us should be worried.
How do real-time facial recognition systems work?
Real-time facial recognition systems come in many forms, but most follow a similar working pattern. Using artificial intelligence (AI), these solutions first detect the presence of a face within an image. Next, real-time facial recognition systems work to analyze the detected face and create a number of relevant data points, which can be used to identify individuals from one another. Finally, the system will compare these data points against pre-existing records and, in turn, should be able to recognize the person in the frame.
What are the benefits of real-time facial recognition systems?
There’s a reason why real-time facial recognition systems are becoming increasingly popular. These solutions offer a much-enhanced layer of security to protect businesses and individuals alike. As opposed to traditional CCTV technology, real-time facial recognition systems enable security measures to be deployed more pre-emptively. For example, these solutions can be used to identify known or suspected threats before a crime is committed, which allows security measures to be proactive, not reactive.
The implications of this benefit alone are massive. With real-time facial recognition systems in place, the threat of major acts of terror occurring can be significantly reduced. Businesses can also use these systems to mitigate the risk of shoplifting. For police and national security services, the technology also offers a powerful tool in the fight to locate wanted or escaped fugitives, who may otherwise be able to avoid detection and commit further criminal acts. Additionally, real-time facial recognition systems can be used to help find missing persons more efficiently.
Why are people concerned about real-time facial recognition systems?
Despite these benefits, there is the potential for real-time facial recognition systems to be used nefariously, particularly if the technology was to fall into the hands of the wrong person, or persons.
These solutions necessitate the creation of databases to store information that can be used to identify individuals. The sheer number of security breaches of major companies and governments over the past year alone should give anyone doubts about whether this information can be properly secured. Unfortunately, if a breach did occur, then criminals would have a powerful new tool for blackmail and extortion.
For example, if a criminal group wanted to breach the security of a particular company, they could use facial recognition software to find the identities of the company’s employees and target them with individualized social engineering or extortion attempts. Since public security cameras are often not secure, it would be quite easy to digitally stalk targets and potentially find compromising information. Sadly, the possibilities for a well-organized and creative criminal group to exploit the technology are endless.
How worried should you be about the technology?
Like with any major technological advancement, real-time facial recognition systems offer both benefits and drawbacks that must be carefully considered as the technology becomes more widely adopted. In the right hands, these solutions provide a more efficient way for businesses, police and security services to mitigate threats and provide an effective tool in preventing crime. However, it’s essential that the appropriate checks and balances are imposed through the sector to ensure that these tools are only ever used for good.
For most of us, real-time facial recognition systems are unlikely to massively alter day-to-day life and should help to make us all safer in the long run. However, it’s imperative that these systems, as well as the databases used to support them, are backed by effective cybersecurity measures. Without this protection, real-time facial recognition systems could be commandeered by criminals and used to target innocent individuals and commit identity fraud on a frightening scale.
Bence Jendruszak is cofounder and COO at SEON.
Welcome to the VentureBeat community!
DataDecisionMakers is where experts, including the technical people doing data work, can share data-related insights and innovation.
If you want to read about cutting-edge ideas and up-to-date information, best practices, and the future of data and data tech, join us at DataDecisionMakers.
You might even consider contributing an article of your own!