The US’s ‘digital verification divide’ is widening; biometrics can help

The digitization of government in the United States is ongoing but closing the so-called digital verification divide poses major challenges, according to a new report.

Based in Washington D.C., the Progressive Policy Institute has reviewed major roadblocks to digitization of government, giving a clear-eyed assessment, while explaining how following NIST’s digital identity guidelines will help overcome obstacles.

The report observes how the digital identity verification divide has been widening, owing in part to “misguided opposition” to biometrics in the digital verification process. This in turn has been slowing down digitization.

A digital verification divide, the report observes, is created by two factors. One is that low-income and marginalized Americans are less likely to have the document trail (bank accounts, mortgages, passports, etc) that people use to establish their identity and authenticate themselves.

The second factor could be considered a perception issue: That the use of biometrics for identity verification has been conflated with the use of biometrics for surveillance and law enforcement. But this conflation is “mistaken,” since the law enforcement approach to facial biometric matching and a typical identity verification system differ. But ongoing debate over the use of biometrics in law enforcement situations has made it harder for policymakers to implement biometrics for identity verification.

The report also makes mention of the issues that have beset “Login.gov,” launched by the General Services Administration (GSA) in 2017. Previously, the capabilities of Login.gov were misrepresented by the GSA, but the portal has received new identity proofing of late.

The report examines the advantages and disadvantages of different approaches to stand-alone verification, such as stand-alone biometrics, which would be highly scalable, with the benefit of low fraud, the report says, but identifies “concerns about biased algorithms” for this particular approach.

In addition, the report points to real-world examples – such as the case of the IRS, and in Puerto Rico – of how passive profiling can lock people out. Instead, the better alternative would be an integrated system that makes use of several approaches to verification, going by NIST guidelines.

Importantly, biometrics are also considered and judged by the report’s author, Michael Mandel, as to whether their use expands or closes the digital verification divide. The review concludes, after analyzing data from ID.me, that biometrics, when part of an integrated verification process, “helps narrow” the digital verification divide.

Identity verification of users is a challenge when faced with a significant population who may have sparse document trails, and an equitable society would regress if these people have trouble verifying themselves in order to access digital government services. It is important then that the goals of inclusion and efficiency “are not in conflict,” according to the report.

In addition, fair treatment of all groups underpins successful government digitization, and for that remote verification may mean “providing an alternative video chat with a ‘trusted referee’ for anyone who chooses, or the equivalent,” the report states.

Finally, the report concludes that closing the digital verification divide can include biometric facial verification using “leading government-tested algorithms,” which can provide a “high level of security and performance” when “done well.”

Article Topics

biometrics  |  digital government  |  digital ID  |  digital inclusion  |  face biometrics  |  fraud prevention  |  ID.me  |  identity verification