West Central Michigan News

The recent struggles of Michigan’s Unemployment Insurance Agency to rise to the challenge of the financial crisis created by COVID-19 has drawn attention to long-known problems in the agency’s computer systems.

The current system is called the Michigan Integrated Data Automated System (MiDAS), and it relies wholly on an algorithm to detect fraudulent claims, according to reporting by Bridge Michigan. 

MiDAS has been documented for flagging claims as fraudulent based on things such as people selecting “I needed the money” from a drop-down menu on the website, as well as groups of applicants with similar birthdays or close geographical proximity.

“MiDAS was programmed to assume people were guilty, and because of that programming, when it’s needed, people aren’t getting their benefits,” Tony Paris, an attorney for the Sugar Law Center in Detroit told Bridge Michigan.

Once claimants have been flagged by the automated system, they may have no other option but to seek to fix the issue by diving into the already overwhelmed phone claims system.

Lawsuits related to the fraud detection system date back to 2013, in which claimants not only faced denial of their claims, but large fines and penalties that were automatically assessed after MiDAS labeled their claims as fraudulent. Bridge Michigan reported that those lawsuits involved tens of thousands of Michigan residents.

Even before the COVID-19 financial crisis, 78% of people who had their claims improperly denied had simply made errors in filling out a claim, which is the highest rate for any state in the U.S.

Some improvements have been made in recent years, such as the requirement for all fraud accusations to be reviewed by a human being.

“It definitely needs an overhaul; there’s no question about that,” UIA Director Steve Gray told Bridge Michigan. “This system seems to be designed with preventing overpayments, and not as much of a priority on making sure that people who are eligible are getting paid.”