It’s not uncommon for consumers to order the same shirt in three sizes and return the two that don’t fit. But what if a consumer exchanges the label and returns another shirt, or buys a dress for an event, wears it once with the tag tucked in, and then returns it?
These are examples of return fraud, and the risks in the retail industry are growing “both in scale and sophistication,” said Juan Hernandez Campos, chief operating officer at Happy Returns, a software company that manages returns for part of UPS. Currently, about 9% of returns are fraudulent, according to the 2025 State of Retail Returns Report, published last October by the National Retail Federation and Happy Returns.
As return volumes increase, fraud remains a persistent problem, Campos said, and merchants can no longer rely on the “highly manual” methods they previously used to process returns.
Instead, retailers and their software partners are turning to AI to help manage reverse logistics, the part of the supply chain that includes processing returns and replenishing inventory. Narvar, which works with companies like DSW and Newell Brands, touts its AI technology that processes billions of consumer data points to detect fraud. Meanwhile, Loop uses AI to automate decision-making and catch fraud in customers like Keene and Princess Polly.
Jackie Swanson, managing partner at Gartner Consulting, said retailers have been transitioning from machine learning-powered fraud detection to AI systems over the past 18 months. the goal? Stop fraud, reduce losses, expedite legitimate returns, and get inventory back on sale as quickly as possible.
Among retailers with more than $1 billion in annual revenue, “almost every retailer is doing something about it,” Swanson said, noting that apparel, beauty and footwear in particular are at the forefront due to high return rates.
AI can combat return fraud
Caroline Reppert, senior director of AI and technology policy at the National Retail Federation, said reverse logistics is “emerging as a key area of focus” as retailers consider how to leverage AI.
Retailers are battling problems with return fraud, including overstating quantities (when consumers claim they returned more items than they actually did), returning empty boxes or sending boxes filled with stones, falsified labels, and counterfeit substitutions. Some consumers go through a wardrobe where they use or wear items for a short period of time before returning them.
According to Campos, one of the big benefits of AI is its ability to analyze large amounts of return data and identify patterns that are difficult to spot manually. For example, Happy Returns, whose customers include Everlane, Pact, and Under Armor, uses technology called Return Vision to detect product discrepancies, such as incorrect logos, altered tags, differences in materials, and product replacements.
When Return Vision discovers one of these issues, it sends that data to the retailer’s dashboard in real time, allowing retailers to quickly see potential problems, Campos said. Retailers can see why an item was flagged for return and see photos of returned items. That way, you have evidence to dispute consumer claims even before the goods arrive at your warehouse.
Happy Returns also uses AI-based behavioral risk scoring to identify potentially high-risk returns based on return frequency, timing, geography, and past behavior. Returns can be flagged for human review before retailers issue refunds.
That said, retailers reveal a complex picture regarding AI in return fraud. In the NRF report, only 45% of companies said they believe AI and machine learning are truly effective in preventing return fraud at their companies.
Bully Max, which makes dog food and supplements, uses Shopify and Chargeflow’s AI systems to detect return fraud, CEO Matthew Kinneman said. In the event of fraud, or if a consumer files a chargeback, the AI automatically generates evidence on behalf of Bully Max and submits it to the credit card company.
That said, Kinneman acknowledged that AI is not perfect. For example, businesses must manually send screenshots to credit card companies. “We found that human review remains essential,” he said.
Faster returns
Fraud is not the only use case for AI in reverse logistics. Bully Max also uses AI to identify patterns, such as repeat reasons for returns. The team can then improve product pages and descriptions to provide consumers with more accurate information before making a purchase.
“In many cases, it’s more valuable to prevent returns than to process them efficiently,” Kinneman said.
Swanson said customers are also using AI to predict returns and make warehousing decisions. In the former, AI models proactively flag orders that are likely to be returned and suggest alternative sizes or products before the consumer finalizes the purchase. In the latter, once a return arrives at a retailer’s warehouse, AI recommends whether the retailer should resell, liquidate, or even destroy the item based on its potential resale value.
Reppert told Business Insider that in the past, returns were done manually, which was time-consuming. AI “allows retailers to process higher return volumes more quickly and consistently.”
Campos said the faster processing also benefits consumers, as refunds for returned items are quickly issued. A positive return experience can build brand loyalty and encourage consumers to purchase from a retailer again, he added.