The Future of Authentication: Are You Embracing Biometrics?
by Katie Björk, Director of Communications and Solution Marketing, Identity and Access Management, HID and Vito Fabbrizio, Managing Director, Biometrics, Extended Access Technologies, HID.
Growing demand for enhanced security that is easier for the end user to manage is pushing the popularity of biometric technology to unprecedented levels, with adoption expected to grow by nearly three times by 2028 to more than $87 billion. Unlike traditional security measures like passwords and physical identification cards or keys, biometric technology uses fingerprints, facial, iris and even behavioural characteristics to authenticate users. The seamless user experience, increased security and ease make biometric authentication a superior option for healthcare organisations, financial institutions, government agencies and organisations needing to improve their cybersecurity strategy.
The appeal and popularity of biometrics are evident. With newer generations of smart devices already coming with embedded biometric authentication capabilities to unlock phones and access apps such as banks, healthcare and digital wallet apps, end users are increasingly expecting these passwordless, contactless experiences.
Today’s Biometrics
An estimated 4.7 billion people use smartphones today, according to Statista, with newer smartphone models coming equipped with at least one integrated feature for biometric authentication, typically involving facial or fingerprint recognition. Biometric authentication presents a secure and convenient substitute for traditional passwords or PINs.
The FIDO passkey is one example. When a user is asked to sign in to an app or website, the user can simply sign in with the same biometric that is used to unlock the device being used (mobile, computer or security key). The passkey’s rapid growth in popularity has some experts predicting it will soon become the dominant form of passwordless identification, with tech giants Google, Microsoft and Apple and key authentication vendors such as HID having already adopted the technology in their ecosystems. In fact, market revenue of passwordless identification systems is projected to grow 27 per cent to 53.6 billion by 2030.
Facial recognition is also gaining popularity, driven in part by government utilisation of driver’s licenses, national ID cards and passports to protect against identity fraud. And, because of advancements in computer graphics, which led to impressive improvements in photo and video quality, facial recognition technology has progressed remarkably. Combined with the latest enhancements in artificial intelligence (AI) and machine learning (ML), for example, facial recognition has become impressively accurate and convenient. While there are concerns regarding matching bias, margins of errors and false match rates, security experts nonetheless say facial recognition use cases will increase significantly over the next few years.
As a result, modern cameras today, when fusing AI and ML with multispectral imaging (MSI) technology, can handle challenging lighting conditions and deliver superior presentation attack detection (PAD) to detect whether there is fraud. Additionally, the possibility to take the algorithms and matching capabilities, which used to only belong in big server databases, can now be done on the camera device, a.k.a biometrics at the edge, so it is able to make biometric processing very quickly, drastically improves the user experience.
Taking Authentication to The Next Level With Behavioural Biometrics
As digital frauds become rampant and more sophisticated, behavioural biometrics also leverage AI and ML to identify unique and measurable patterns in human behaviour. Behavioural biometric data consists of identifying and learning actual user activities so that identity is validated and intent is understood.
User activities such as logging in to the application, navigation to a specific page, transaction checkout, etc. and the data that characterises the user as a human being, such as mouse movements, typing cadence, touch events, swipe patterns, etc., represent behavioural biometry, which can uniquely characterise a user with a reasonable amount of analysis and proper processing.
This profile is continuously updated, improved and enriched in real-time, so it can be used to identify a user uniquely and seamlessly. These markers, augmented with user-specific information such as IP address, device, time and navigation patterns, deliver the best contextual value resulting in a high fraud detection rate and decreased number of falsely rejected users.
The Paradox of Regulatory Compliance And Trust
Regulatory concerns go hand-in-hand with technological advances and biometric identification is no exception. Privacy laws are expanding in tandem with biometrics on a global scale, including the European Union’s General Data Protection Regulation (GDPR), the California Consumer Privacy Act (CCPA), Health Insurance Portability and Accountability Act (HIPAA) and others.
In an effort to accelerate Malaysia’s digitalisation initiatives, the Digital Identity Development Project (IDN) was spearheaded to enable a user-controlled biometric digital identification and verification system that matches an individual’s biometric data with a government database, to distribute targeted subsidies and other benefits to the populace, amongst other uses.
It is unquestionable that protecting an individual’s biometric data, and making sure companies use the data responsibly is important. But the paradox is that this very biometric information is used to protect the integrity and the privacy of someone’s identification, their assets and the access to places.
The complex regulatory landscape can be a barrier to the broader adoption of biometric technologies. Many smaller organisations are fearful of some of the laws and how they have been enacted and enforced by expensive lawsuits and fines.
There are other challenges impeding the broader adoption of biometric technology. One is the public’s worry that their biometric data will be secure and ethically managed. The second is consistency with the devices it is being used to secure. Biometrics are highly dependent on the environment in which they are captured—the device in broader terms. If it is on a camera, that camera and the reader must work reliably.
It is also important to consider the suitability. For example, requiring facial or fingerprint recognition is not ideal when the user is driving a car. In this case, voice recognition would be a better fit.
Best Practices for Growing Adoption
Education and communicating with customers and end users will help them better understand how the biometric technology being used works and what will happen with the data. For example, employers using biometrics for time and attendance or at the point of sale for fraud and abuse prevention should explain how the technology is helping the customer, what is being collected and why.
To this end, it is important to partner with companies that have a modular approach to biometrics, as there is no one-size-fits-all solution. In other words, trying to deploy the same solution for every use case or industry is not ideal. Instead, it is important to stay laser-focused on the end-user experience by reducing the level of friction, using data encryption, obtaining user consent and creating different components to allow resellers and end users to integrate biometrics into their applications as seamlessly as possible. This way, a solution can be tailored to the various use cases and maybe even their geographic regions.
Looking Forward
The future of biometrics is bright as the public’s comfort level with the technology’s security grows.
As advancements in biometric technology continue to evolve and security providers adopt ethical and privacy considerations, as well as address all potential vulnerabilities within their solutions, the public’s trust and confidence will continue to strengthen.
