The many challenges with tax refund fraud detection

by Guest Contributor 3 min read October 8, 2013

Billions of dollars are being issued in fraudulent refunds at the state and federal level. Most of the fraud can be categorized around identity theft. An example of this type of fraud may include fraudsters acquiring the Personal Identifying Information (PII) from a deceased individual, buying it from someone not filing or otherwise stealing it from legitimate sources like a doctor’s office. The PII is then used to fill out tax returns, add fraudulent income information and request bogus deductions.

Additional forms of tax refund fraud may include:

  • Direct consumer tax refund fraud using real PII of US Citizens to file fraudulent tax returns and claim bogus deductions thereby increasing refund amounts
  • EITC (Earned Income Tax Credit)/ACC (Additional Childcare Credit) fraud which is usually perpetrated with the assistance of a tax preparer and claiming improper cash payments and/or deductions for non-existent children.
  • Tax Preparer Fraud where tax preparers purposefully submit false information on tax returns or file false returns for clients.
  • Under reporting of income on tax filings.
  • Taking multiple Homestead Exemptions for tax credit.
  • Since this Fraud more often occurs as an early filing using Fraudulent or stolen PII the individual consumer is at risk for long term Identity issues.

Exacerbating the tax refund fraud problem:

  1. The majority of returns that request refunds are now filed online (83% of all federal filings in 2012 were online) -if you file online, there is no need to submit a W-2 form with that online filing. If your employment information cannot be pulled into the forms by your tax software you can fill it in manually. The accuracy of information regarding employer and wage information for which deductions are based, is only verified after the refund is issued.
  2. Refunds directly deposited – filers now have the option to have their refunds deposited into a bank account for faster receipt. Once these funds are deposited and withdrawn there is no way to trace where the funds have gone.
  3. Refunds provided on debit cards – filers can request their refund in the form of a debit card. This is an even bigger problem than bank account deposits because once issued, there is no way to trace who uses a debit card and for what purpose.

So what do you need to look for when reviewing tax fraud prevention tools?

  • Look for a provider that has experience in working with state and federal government agencies. Proven expertise in this domain is critical, and experience here means that the provider has cleared the disciplined review process that the government requires for businesses they do business with.
  • Look for providers with relevant certifications for authentication services, such as the Kantara Identity Assurance Framework for levels of identity assurance.
  • Look for providers that can authenticate users by verifying the device they’re using to access your applications. With over 80% of tax filings occurring online, it is critical that any identity proofing strategy also allows for the capability to verify the source or device used to access these applications. Since tax fraudsters don’t limit their use of stolen IDs to tax fraud and may also use them to perpetrate other financial crimes such as opening lines of credit – you need to be looking at all avenues of fraudulent activity
  • If fraud is detected and stopped, consider using a provider that can offer post fraud mitigation processes for your customers/potential victims.

Getting tax refunds and other government benefits into the right hands of their recipients is important to everyone involved. Since tax refund fraud detection is a moving target, it’s buyer beware if you hitch your detection efforts to a provider that has not proven their expertise in this unique space.

Related Posts

Customer Spotlight: How Matrix Rental Solutions Strengthens Trust in Affordable Housing

Learn how Matrix continues to deliver a secure, trusted rental experience as fraud tactics evolve. Read more!

July 31, 2026 by Laura Burrows
What Is AI Decisioning?

Every business makes decisions about people and transactions all day long. Should we approve this loan? Is this purchase fraud? Which customer should get this offer, and what should it be? For a long time, those decisions were made in one of two ways: a person reviewed each case by hand, or the company wrote fixed rules, like "approve anyone with a credit score above 700." Both work. Both also leave value on the table. The manual review is slow and hard to scale. The fixed rule can turn away good applicants and is slow to adapt when the market shifts. AI decisioning is a third way. What makes AI decisioning work Instead of relying on a single reviewer or a rigid rule, automated decisioning uses models that learn from data — studying how thousands of past cases turned out, finding the patterns that predict an outcome, and applying them to each new decision, often in real time. The result is faster, more consistent decisions. But a model on its own isn't the whole story. Getting real value from AI decisioning takes good data to learn from, AI analytics to generate insights, the tools to act on it and the governance to keep it compliant. What we've found is that the pieces only pay off when they work together, and that is where we're built differently. A model is only as good as what it learns from, and we pair your data with one of the deepest views of consumer and commercial credit: decades of full-file history and vetted attributes. Then we give you the tools to act on it. Use cases across your business Whether you're trying to grow your customer base, reduce fraud, manage lending risk, or improve collections, automated decisioning brings all the pieces together to make more accurate, consistent and explainable decisions at scale. Fraud and Identity A fraudulent transaction that slips through costs money and erodes trust. Rules are static, and fraudsters move fast. They'll probe boundaries, find the blind spots and move to the next scheme. By the time the rules are updated, they're already three steps ahead. How AI decisioning changes this: AI fraud detection with real-time risk scoring and decisioning across transactions and customer interactions Intelligence that continuously learns from results to help adapt fraud strategies as threats evolve Reduced false positives and less friction for customers at account opening and checkout Identity verification tools that confirm someone is who they say they are without slowing down the experience Credit and Lending Loan approval is where the relationship begins. Credit risk decisioning helps lenders find that delicate balance between approving enough people to grow, but carefully enough to manage risk. Missing that balance means turning away good customers or taking on losses that are difficult to absorb. How AI decisioning changes this: Increased approval opportunities for creditworthy applicants without increasing overall risk Models you can update and deploy quickly as market conditions change, rather than waiting months Ability to run "what-if" scenarios to test how a new strategy would have performed on your historical data before putting it live Collections Which customer should your team reach out to today? Through which channel? What kind of message? If you reach out too aggressively, you push someone who might have recovered into default. If you wait too long, you lose them. If you call someone at work, they resent you; if you text, they might ignore it. If you offer a payment plan, they might accept it, but only if the terms make sense to their financial situation. How AI decisioning changes this: Optimized next-best-action and contact-channel strategies for each individual customer Improved recovery potential through better targeting Less time spent on accounts with a lower propensity to pay, freeing your team for higher-impact cases Ability to segment and test new strategies before rollout Customer Acqusition Finding the right customers is about reaching the right people with the right offer at the right time. To stay competitive, it’s now a requirement to balance growth with risk while creating a seamless experience converting prospects into customers. How AI decisioning changes this: More precise prospect targeting using credit, behavioral, and alternative data, where permitted, to identify consumers most likely to respond Personalized offers delivered in real time Dynamic decision strategies that can be updated quickly as market conditions and customer behavior change Ongoing testing and optimization of acquisition strategies to improve campaign performance and support customer lifetime value Driving results with AI decisioning Every customer interaction is a decision. Businesses that can adapt quickly will be better positioned to grow, manage risk, and deliver the experiences customers expect. The technology will continue to evolve, but the goal remains the same: making informed decisions that balance business objectives, risk, and customer experience. Learn more about our decisioning software

July 27, 2026 by Zohreen Ismail
Why Innovation Matters for Members First Credit Union

Learn how Members First Credit Union uses innovation and data-driven insights to better serve members and expand financial opportunity.

July 24, 2026 by Scarlet Nickel

Subscribe to our Newsletter

Enter your name and email for the latest updates.

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.

Subscribe to our Newsletter

Don't miss out on the latest industry trends and insights!
Subscribe