How Secure Is Facial Recognition Technology?

Posted by Kelvin Chandran on 2/21/19 10:30 AM
Kelvin Chandran
Find me on:

How Secure Is Facial Recognition Technology?

Facial recognition and other biometrics technologies have gained massive adoption from a wide range of companies in the past few years. In this article, we'll examine how these technologies work. Then, we'll look at some current real-world applications as well as the risks and benefits for individuals and organisations.

What Is Biometrics?

Biometrics is the science of measuring physical and/or behavioural characteristics that are unique to each individual in order to verify an individual’s identity. Numerous types of biometric data types are in use today. A few examples include retina, fingerprint, and facial recognition. In this article, we focus specifically on facial recognition.

How Does Facial Recognition Work?

Just a few years ago, most security systems relied upon rather unsophisticated photography and videos to verify an individual’s identity. These technologies were typically only effective if another person recognised a particular individual, making these systems mostly insufficient for security purposes.

Facial recognition research has been going on since the 1960s, but automated facial recognition technology has actually only been relevant in real-world applications since around 2009. Moreover, it has only gained major momentum in its adoption in the past couple of years. Generally, facial recognition is used as a security system technology. It works by comparing the features of a given person’s face with an existing image stored in a database. It either validates or denies access to services and applications based on matching data or lack thereof.

Current Consumer Applications

Possibly the most commonly-used application today for facial recognition technology is for accessing electronic devices such as phones, laptops, and tablets. For example, the iPhone X allows users to unlock their devices via a built-in Face ID. This is a biometric data feature that builds upon the previously-released Touch ID, an unlock feature that uses fingerprint matching technology.

In 2019, facial recognition appears to be the biometric trend of the future with Apple Pay, the App Store, iTunes, and various third-party iOS applications now all allowing users to login via Face ID. But just how effective is this current application in its efforts in providing a real-world security solution? According to Apple, the risk that someone is able to login to another person’s device with Face ID is only 1 in 1,000,000. This is a 20x improvement in security when compared to the 1 in 50,000 odds with Touch ID.

Facial recognition is also gaining adoption in a number of other consumer and government applications, especially in China. For example, KPro (KFC’s brand in China) teamed up with search engine giant Baidu and payment processing giant Alipay, to allow customers to pay for items simply by smiling in front of an in-store camera. There’s also the city of Shenzhen, which uses facial recognition technology to identify jaywalkers on CCTV footage and send automated fines.

Potential Risks and Technical Challenges

Privacy Concerns and Public Opinion:

One of the biggest concerns that individuals have with biometric technologies like facial recognition is the potential impact for invasion of privacy. On one side, these technologies do have the potential to increase security and improve accessibility. However, there are also legitimate concerns that this technology could be exploited by companies that could sell sensitive user data.

For example, a store could theoretically use facial recognition to know how many times you have visited and/or combine this information with purchase data to conduct targeted advertising campaigns. 7-Eleven has already begun implementing such technologies in 11,000 of its stores in Asia. Another big concern that some people have is the potential that this data can be accessed by hackers. As a result, hackers could find new ways to use biometric data for criminal activities like identity theft. According to Iris ID Systems Inc., this is already a real-world issue in India as 1.3 billion citizens now have biometric identities.

Statistics from the Center for Data Innovation, however, show that the public actually isn’t opposed to facial recognition technology. Surveyees in the US were asked “Agree or disagree? The government should strictly limit the use of facial recognition technology even if it comes at the expense of public safety?” Only 18.3 percent agreed. A staggering 54.8 percent disagreed. In a number of other hypothetical scenarios, respondents also said that they would rather choose security over privacy.

Legal and Business Risks:

For organisations, implementing such technologies also comes with some risks. Facial recognition leaders like Apple have to recognise the possible legal implications of implementing this technology. Even if it is possible to gain regulatory approval in the US, GDPR policies in the EU provide an example of higher privacy standards globally. In many locations around the world, applying facial recognition could pose more risks than potential rewards for businesses.

In an area where few existing regulations exist, some companies are hesitant to spend time and resources. While being cautious might be a good idea in some scenarios, there is also the possibility that doing so will put organisations at a competitive disadvantage vs. others that do pursue R&D and implementation of facial recognition technology.

Biometrics: Improving Security for Individuals and Organisations

It’s evident that organisations are moving towards the use of biometric data in a variety of applications. Yes, potential risks do exist. However, it is certainly possible to reduce or eliminate these risks and provide better security for end users. For example, storing biometric data on the blockchain or other other types of decentralised databases could help ensure top-notch data security.

In other words, blockchain vastly improves upon traditional data storage technologies that are vulnerable to security threats. Additionally, by creating applications that allow users to have full control over deciding which parties can access biometric data, individuals are able to stop unknown third parties and take an active role in the protection of their own biometric data.

For organisations that require identity checks, facial recognition and other biometric technologies can provide a more effective solution for increasing the accuracy of identity verification. These solutions can also help fight common issues like identity theft as well as ensure compliance to a wide range of identity-related policies (i.e. KYC, AML, CFT, and more). For individuals, this can mean easier signup or login processes that save time. This gives individuals a simplified more secure option for gaining access to a variety of services via biometric verification.

Topics: Data ownership, Digital identity

Recent Posts

Subscribe here