Not that long ago if someone wanted to commit receipt fraud, they would need at the very least an online template, if not a full education in photo manipulation and editing. Not anymore. In an extremely short time, AI has become the primary way people commit the act.
This is according to data from two recent studies, one from AP and expense management solutions provider
AppZen found the proportion of AI-related fake receipt flags on their platform went from zero to 70.8% in just over a year. This does not mean AI-guided expense fraud grew by that proportion but, rather, when a receipt is identified as fraudulent, it is now almost always AI-generated.
Across AppZen's enterprise platform, which flags receipts suspected of being fraudulent, 0% were AI-generated in March 2025. By the middle of May this year, however, that proportion had soared to 70.8%. This is based on 1,471 AI-generated fake receipts, submitted by 745 employees at 174 companies, claiming a total of $148,143 in fabricated expense reimbursement.
Previously, receipt fraud was dominated by templates from specialized websites and small online services. Just over a year ago, such template-based fakes accounted for 95% to 100% of the receipts caught by AppZen's "fake receipt check" rule. By May 2026, they made up just 29% of the total. The crossover happened in April 2026. At this point, AI image generators are now the dominant fabrication tool.
This was an extremely sudden jump. In the second quarter of 2025, AI-generated take receipts accounted for 26% of flagged items. It dropped to 11% in Q4 2025, grew slightly to 13% in Q4 2026, then in April 2026 exploded to 39% and then, less than one month later, soared to 71%.
The shift has affected not just the way people commit fraud but for how much. The report said per-receipt dollar size is remarkably consistent across regions, with an average of approximately $100 in every environment, and a median of approximately $32. In comparison, fake-template-website fakes (the older method this replaces) averaged $182 each, almost twice as large. Overall, AI-generated expense fraud is a high-volume, low-dollar attack pattern, deliberately sized to slip under most companies' auto-approval thresholds. They may be smaller, but they're more numerous.
The platform data lines up neatly with the survey data from Emburse, which conducted a poll of 2,000 workers in the U.S. and U.K. They found that 34% of respondents admit they have used AI to generate a fake receipt for a business expense. U.S. respondents reported the behavior at a higher rate, with 40% saying they have done so, compared with 29% of U.K. respondents.
Furthermore, 40% say they used AI tools funded by their own employer to generate those receipts, while 9% have built their own AI tools to generate fake receipts.
More specifically, Emburse found that 19% have used AI to completely fabricate a purchase, 15% to increase the value of an item they purchased, and 6% have used AI to replace a receipt they lost for a genuine purchase.
"AI makes it easier than ever to fabricate receipts and bypass traditional controls, so organizations need a more proactive, AI-centric approach to managing spend," said Michele Shepard, chief revenue officer of Emburse. "With AI increasingly used to create convincing fake receipts and manipulate expense data, the only effective defense is AI-powered prevention and detection. Organizations that aren't leveraging AI will be at a significant disadvantage in finding and stopping these emerging threats. Identifying expense fraud and abuse remains critical, but the most effective strategy is preventing non-compliant spend before it happens through clear policies, fast reimbursements, virtual cards and AI-powered controls that improve the employee experience while increasing visibility for finance teams."
As for the reason behind this sudden increase, at least part of the blame can be laid on the general economic condition. When asked whether they had ever made a personal purchase and submitted it as a business expense because of their financial circumstances, 22% of respondents overall said yes. Another 22% said they have considered doing so but ultimately chose not to. The poll noted that 76% of U.S. respondents and 74% of U.K. respondents said they were somewhat or very concerned about their personal finances.
Many do not seem to feel guilty about this, viewing the behavior as not especially harmful to their employer and, in some cases, they said it makes them a better worker. The poll found that faking receipts ultimately makes them more effective at work, with 28% saying it helps them learn AI faster, and 16% saying they are simply making use of large, underutilized token budgets. But beyond this, a sort of malice is also playing a role. The poll found that 18% of the respondents are motivated by fear or revenge, reporting that they're taking advantage of AI tools for their personal lives because they believe employers will replace their role with AI. Another 22% are using AI tools paid for by their employer to apply for other jobs.
Based on Emburse's historical data, this trend appears to be growing, especially among older workers aged 55 and up; those who admitted to "revenge spending" among this group went from 4% in 2024 to 11% in 2026.
However, another factor might be the reimbursement process itself and the attendant financial pressures. Emburse found that 51% of U.S. respondents and 45% of U.K. respondents incurred overdraft fees, late-payment charges, excess fees or credit card interest while waiting for reimbursement; 29% of U.S. respondents and 21% of U.K. respondents incurred interest on personal credit cards because reimbursement took too long; and 23% of U.S. respondents and 16% of U.K. respondents delayed personal purchases or bill payments while waiting to be reimbursed.
"For the most part, employees aren't intentionally misusing corporate AI tools," Shepard said. "In many cases, they've been encouraged to adopt AI quickly and integrate it into their daily lives. The real challenge for employers isn't bad behavior, it's a lack of visibility. When employees use company-funded AI for everything from personal tasks to job searches, organizations need a clearer understanding of what's being purchased, how it's being used, and whether it's delivering value. As AI spending continues to grow, finance and technology leaders will need the same level of oversight and accountability they've long expected for every other category of business spend."
While AI-generated fake receipts are relatively new, they are
