ID.me unemployment benefit facial recognition

April 2021
Updated: February 2022

People across the US are being denied unemployment benefit payments due to problems with a facial recognition system that is supposed to prevent fraud.

According to VICE/Motherboard, unemployment beneficiaries in 22 US states have been experiencing issues with identity verification service ID.me's facial recognition system for months.

The issue has led to people being locked out of their unemployment accounts and not receiving the funds they are due. Many have taken to airing their complaints in public as they have been unable to contact ID.me support staff.

ID.me CEO Blake Hall suggested to VICE that user error may be to blame, and that the company was not aware of 'eligible individuals' who had been unable to verify their identity.

IRS mandatory facial recognition

In November 2021, the US Inland Revenue Service (IRS) announced it will require taxpayers to use a selfie to verify their identity with ID.me before using some of the agency's online services.

Others voiced their concerns that facial information collected by the IRS collects could be reused without users’ knowledge. The company's service terms enable the company to share people’s data with the police, government, and 'select partners', stoking privacy concerns.

Early 2022, security researcher Brian Krebs highlighted a number of security issues with ID.me's identity verification system. Blake Hall later admitted in a LinkedIn post that ID.me uses a one-to-many facial recognition system that searches for individuals across multiple databases.

On February 7, 2022, the IRS announced it will stop using ID.me for verification purposes. In May 2022, a group of Democratic lawmakers urged the US Federal Trade Commission to investigate ID.me, claiming (pdf) that its CEO made misleading comments about how the company uses facial recognition.

Operator: ID.me; US Inland Revenue Service (IRS)
Developer: ID.me
Country: USA
Sector: Govt - welfare; Govt - tax
Purpose: Verify identity; Detect fraud
Technology: Facial recognition
Issue: Accuracy/reliability; Privacy; Bias/discrimination - race, ethnicity
Opacity: Complaints/appeals

Reference

Research, audits, investigations, complaints

News, commentary, analysis