The UI Fraud Landscape in 2026

Optimizing Investigative Resources in an Era of Infinite Data

This is the first post in our latest series, tracing the UI fraud lifecycle from detection through recovery.

From Claims Processing to Fund Stewardship

The mandate for state workforce agencies has changed. A decade ago, the job was largely operational: take in claims, verify eligibility, issue payments, move on. That job still exists, but it’s no longer the whole picture. Trust funds are under sustained pressure, from economic cycles that drive claim volume up and down unpredictably, and from fraud schemes that have grown more organized and more automated. Agencies are being asked to do something closer to fund stewardship: protect the money, not just process the paperwork.

That distinction matters because the two jobs call for different defenses. Claims processing is built around throughput, getting eligible claimants paid quickly and accurately. Fund stewardship is built around exposure, understanding where money can leave the system improperly and closing those paths before it happens. Most agencies are still running processing-era systems against a stewardship-era threat, and the gap between the two is where losses accumulate.

For every dollar an agency recovers after a fraudulent payment has already gone out, several more are lost to schemes that were never caught, or caught too late to matter. That ratio is the real cost of a reactive posture, and it’s the reason detection has to move earlier in the process, not just get faster once a case is already open. It also reframes how agencies should think about technology investment: the highest-value dollar isn’t the one spent speeding up recovery, it’s the one spent preventing the loss in the first place.

Ghost Employers Haven’t Gone Away

Fictitious employer fraud remains the most expensive version of this problem, and it hasn’t slowed down. Unlike a single claimant misreporting earnings, a fictitious employer scheme creates an entire shell company that exists only on paper. Once that entity is established in the tax system, it can be used to generate synthetic wage records for equally fictitious “employees,” who then file unemployment claims against wages that were never actually earned or paid.

These schemes have gotten harder to catch precisely because they’ve gotten more organized. Rings of fraudulent employer accounts, often using synthetic or stolen identities, can be stood up faster than manual review processes can flag them. Left unchecked, the damage doesn’t stop at the fraudulent payments themselves. Legitimate employers absorb part of the cost too, since fraudulent charges can distort the experience ratings used to calculate their own UI tax rates. It’s a quiet tax on honest businesses, on top of the direct hit to the trust fund.

This isn’t a hypothetical risk. Wyoming’s Department of Workforce Services recently partnered with Catalis specifically to target fictitious employer schemes, a reflection of how seriously states are now treating this exposure. We covered the mechanics of these schemes in more depth in our original look at fictitious employer fraud prevention, which remains a useful primer if you’re newer to the topic.

The pattern holds across states: the schemes that do the most damage are the ones built to look ordinary. A shell company that never files a single unusual document is far harder to catch through manual review than one that raises an obvious flag, which is exactly why relying on staff to notice anomalies by eye stops scaling once fraud rings professionalize.

Pre-Payment First, Still the Right Philosophy

The most effective response isn’t a bigger investigations team working faster after the fact. It’s catching the pattern before the first payment goes out, which means breaking down a divide that’s existed in most agencies for years: employer tax records and claimant systems operating as two separate worlds instead of one connected view.

When those systems talk to each other, the behavioral pattern behind a fictitious employer scheme becomes visible early. How was the employer account established? How quickly did claims start coming in against it, and do the claimant details line up with a real business operating in that industry and location? Agencies that can ask those questions automatically, before funds move, catch far more than agencies asking them manually, after the fact.

That’s the philosophy running underneath this entire series. Detection, determination, and recovery aren’t three separate problems handled by three separate teams working from three separate systems. They’re stages in one lifecycle, and the agencies making the most progress against fraud are the ones treating them that way. Over the next few posts, we’ll walk through what that looks like in practice: how leading agencies are combining cross-matching, SIDES, and automated tip intake to catch fraud earlier, what a defensible determination process looks like once a case is confirmed, and how automation is finally closing the loop on overpayment recovery.