This is a guest post written by Sarah Amundsson, senior business developer at Shufti Pro.
Identity theft is one the fastest growing crime in the US with the identities of millions of individuals compromised in recent years. In 2019, the Federal Trade Commission received 3.2 million reports of consumer protection issues. Out of these, the most common fraud reported in the last year was imposter scams and identity theft, especially in government services, online shopping and credit bureaus.
With an increase in massive and big-scale data breaches, traditional verification and authentication methods have been compromised. For instance, 5 million customer’s data was stolen by hackers in a recent DoorDash data breach. The personal information including financial details and passwords was compromised and is prone to identity theft. Data breaches like these resulted in many cases of identity theft.
This increase in data breaches contributed to a 120% increase in identity theft and account take over frauds. Account takeover fraud is being fueled by this change in the identities and information stolen by hackers. It makes the digital onboarding of the customers a risky affair for the businesses and due to these threats businesses are adopting more advanced solutions to deal with these issues. Many online businesses are increasingly adopting AI-powered solutions that require users to verify their identities.
With this wave, the old and traditional authentication methods are being replaced by biometric verification. Biometrics uses an individual’s unique biological traits to verify his or her identity. This inherence based identification has more potential than the possession and knowledge-based methods for authentication. Nevertheless, biometric authentication is also prone to attacks. Most observed attacks include presentation attacks such as spoofing, printed attack, deep fakes, 3D mask attacks, and makeup attacks.
Governments and private businesses are adopting mobile biometrics to enhance the security and customer experience by speeding up the verification process in financial services, military, public transportation, and healthcare industry as well. It is imperative to deter and take effective countermeasures to address these attacks.
Fighting sophisticated presentation attacks
To address presentation attacks, the field of biometric security has attracted a lot of attention.
Deepfakes and spoof attacks with liveness detection
Deepfakes intend to spread misinformation, however, another major concern with the manipulation of personal audio, video and other digital footprints could have a massive effect on an individual level. As more and more data is personal data is shared online, deepfakes will become more problematic for those susceptible to identity theft.
While attackers can use deepfake techniques to convincingly imitate the likeness of an individual, it is still difficult to digitally impersonate someone’s voice without fairly obvious imperfections, according to Robert Capps, VP of Market Innovation for NuData Security, a Mastercard company. He further added that deepfakes audio or video cannot currently be rendered in real-time, without an attacker having a large volume of computing resources and a lot of high-quality audio/video source material to train computer machine learning algorithms.
While deepfakes can be convincing to other humans, they are unable to pass physical or passive biometric verification, so coupling strong liveness detection, along with the collection of passive and physical biometric signals to verify a user’s identity, largely mitigate the current risks presented in banking transactions.
AI-backed liveness detection
As it sounds, liveness detection determines whether the face presented to a facial recognition system is that of a live person, a 3D mask photo, or a cut-out photo. Based on the presentation attack algorithm (PAD), liveness detection helps in preventing any spoof attacks by the criminals.
Facial liveness check is a crucial step for accurately authenticating the person with facial verification technology in the automated verification scenarios such as mobile authentication, online customer onboarding and authenticating any online transaction. Liveness detection technology uses machine learning algorithms trained on large data sets to identify presentation attacks.
With the proliferation of facial technology, fraudsters have become active. Fraudsters use different spoofing techniques to attempt unauthorized access. As businesses rely heavily on digital onboarding, there’s a need to introduce advanced AI-powered facial recognition technology that could help fight against criminals and enhancing facial recognition with 3D liveness detection provides a foolproof security solution.
About the author
Sarah Amundsson is a senior business developer at Shufti Pro. Sarah leads customer-facing teams and plays a key role in driving customer goals and product utilization. As an expert in digital identity verification, she helps businesses deploy solutions globally to solve KYC, KYB and AML problems for both for individuals and businesses.
DISCLAIMER: BiometricUpdate.com blogs are submitted content. The views expressed in this blog are that of the author, and don’t necessarily reflect the views of BiometricUpdate.com.
3D | authentication | biometric identification | biometrics | deepfakes | facial recognition | fraud prevention | identity verification | liveness detection | Shufti Pro