shopping with facial recognition

What it Takes to Create Trusted Identities in a Digital Economy

Acuant CEO, Yossi Zekri, was interviewed in this exclusive with The Paypers. You can read the original here. Acuant was also a featured solutions provider in The Paypers Fraud Prevention and Online Authentication Report 2019/2020.

 

If we look at the current challenges faced in the identity-based transactions ecosystem, what risk issues are the most prevalent and why?

We all see the steady headlines regarding major data breaches and hackings that have penetrated leading corporations, which has yielded significant fake identities. Additionally, the rapid pace at which we have become a mobile-first world is now presenting new fraud challenges that businesses must quickly address.

AI-based identity fraud, including deepfakes and video replay, is also growing. Elementary identity proofing including passwords, social security numbers, and knowledge-based authentication are no longer viable for businesses who want to meet compliance regulations such as KYC and AML.

Synthetic identity fraud is another one to watch, and according to McKinsey, it is the fastest growing type of financial crime in the US, accounting for 80% of all credit card fraud losses. With synthetic ID fraud, there is no person to trace or collect from, so banks and lenders must take steps to prevent it. Financial institutions should look beyond a credit score and bring in third-party data such as social media accounts, cell phone records, previous addresses, email addresses, and phone numbers, while seeking a high degree of consistency for those attributes and ensure they are tied to a proven identity.

What must companies keep in mind when employing identity verification solutions?

There are no global standards today in identity verification, but there are solutions that offer a level of certainty for every level of risk. It is up to institutions to decide the amount of security they are requiring, and up to consumers to decide how much friction they are willing to bear. Consumers are more willing to bear it for things like opening a bank account as opposed to making a low value purchase on an app.

Establishing a trusted identity is the base for the digital identity of the future — and this is at the heart of what Acuant does. We believe you must start with document authentication to establish a trust anchor to set a level of trust associated with the initial identity/person. Afterwards, you can layer on specific solutions to further perform additional identity proofing (e.g. biometrics, device fingerprints, behavioural, data points etc.).

What role does the remote identity verification play in improving customer experience? Moreover, what are the best methods for verifying an identity remotely?

Being able to transact from anywhere at any time in today’s ‘On Demand’ culture has led to opportunities that can increase revenue, but also led to increased fraud attempts.

Verifying identities remotely creates additional challenges for businesses. They must be able to establish genuine presence, taking into account users’ risk appetite, and offer a diversity of solutions based on the type of transaction and the problem being solved. Once a trusted digital identity has been established, it paves the way for it to be re-used with less friction and increasing trust.

We recommend the use of multi-factor authentication, which can be quite swift and seamless today. The best methods start with capturing and authenticating a quality image of an identity document and then binding it to the person using the document with some form of biometric such as facial recognition with strong liveness test. For remote transactions this can be done via desktop or mobile device. Image capture features such as auto-cropping and glare detection can not only boost user experience, but also significanlty improve results 100%. Solutions must also be in real time, to decrease abandonment rates.

What is an identity corroboration hub?

An identity corroboration hub is the ability to build and a true digital identity. This means that you have established a verified identity via a trust anchor, tied it to a person, and now have access to a trusted identity platform that continuously authenticates this identity/person through real-time monitoring. Gartner has stated that by 2023, identity corroboration hubs will displace existing authentication platforms in more than 50% of large and global enterprises. They also state that there is a growing trend toward orchestrating identity proofing, fraud detection and orthodox credential-based methods that are traditionally used to provide confidence in the identity of returning customers.

The corroboration hub offers tracking and analysis of true identities linking their connections and their activities in real time, while protecting their privacy. This allows for true identity verification at any point in time, giving the ability to pinpoint fraud and also decide when and how you want to verify an identity (with what data points), expanding the scope of identity proofing.

Identity corroboration hubs prevent new account and transaction fraud, account takeovers, identifying the use of stolen or synthetic identities, and enable compliance with regulations such as KYC and AML.

How do AI and machine learning power up the current strategies of customer verification?

Is there any room left for manual checks in the future? Computer software can independently process millions of transactions at a rate unachievable by human experts and is necessary to scale. However, humans can apply their expertise and teach the computer how to spot the issues in the future. Human-assisted machine learning allows the model to improve through constant input and refinement of data with the oversight of a trusted identity expert.

This marriage of AI and human experts passes more good IDs/ customers, while catching sophisticated fakes, and providing a seamless customer experience. The key is to start with an automated solution that uses a strong algorithm, have a relevant data set, then add human oversight. And for cases where 100% certainty is needed or automated review was not able to provide an adequate answer for transaction, there is still manual review of documents by trained forensic experts.