Return Abuse Goes Mainstream as Nearly Half of Consumers Use AI in Return Claims, New Riskified Research Finds

Riskified , a global leader in ecommerce fraud detection and risk intelligence, released a new global report, “Rewriting the Rules on Returns”, exploring how consumer attitudes and behaviors around ecommerce returns are evolving in the age of artificial intelligence (AI).

The Riskified-commissioned study, conducted by eTail Insights, is based on a survey of 2,091 consumers across seven countries, alongside in-depth interviews with senior leaders from many of the largest retail companies in the world. The research finds that return abuse behaviors are increasingly normalized, while nearly half of consumers already use generative AI tools to assist with return or refund claims. At the same time, merchants are responding by tightening return policies, shortening return windows, and deploying advanced AI detection to better distinguish between legitimate behavior and abuse patterns.

Read More: SalesTechStar Interview with Ilyas Kurklu, Co-founder and CEO of Replenit

Key Findings

Nearly half of consumers (50%) report using generative AI tools such as ChatGPT or Claude to help draft return or refund claims, signaling a structural shift in how consumers interact with post-purchase processes.

Retail leaders interviewed for the study note that shoppers are using AI to write highly convincing return requests, which is overwhelming retailers. At the same time, the research shows that many shoppers view a range of return-related behaviors as acceptable:

• 46% think it’s fine to return something just because it looks different in person than it did online
• 42% think it’s okay to buy multiple sizes or colors to try on at home, fully intending to return most of them (aka “bracketing”)
• 24% think it’s acceptable to wear or use an item before returning it for a refund.

Social media plays a role: More than half of consumers report encountering return-related content on social media, where “life hacks” and informal guidance around returns are increasingly shared and normalized. Overall, 85% of consumers accept at least one type of strategic or borderline return behavior, while only 15% prefer to stick to the policies.

In response, many retailers are tightening return policies, while some are also beginning to introduce more tailored approaches based on customer risk and behavior, though these are not yet consistently applied at scale. Common changes include narrowing return reason categories, shortening return windows, offering store credit instead of refunds, strengthening inspection and quality control processes, and applying differentiated treatment for high-risk versus low-risk customers.

Read More: You Have Cloned Your Voice. Now Your AI Is Making Cold Calls

Consumers are not uniformly opposed to stricter controls. Surprisingly, the majority (52%) are supportive of stricter return policies, with more than 60% saying they would adjust their behavior if they better understood the true cost of returns. More than half (56%) also prefer personalized or tiered return policies rather than uniform approaches.

“The democratization of generative AI has fundamentally changed the returns landscape. When half of consumers are using AI to draft highly persuasive refund claims, manual reviews simply cannot keep up,” said Jeff Otto, Chief Marketing Officer at Riskified. “To protect margins without alienating top shoppers, retailers need the ability to provide differentiated return experiences, accurately and at scale. Using AI and identity-based intelligence, retailers can dynamically reward loyal customers with a frictionless process while automatically applying strict controls to policy abusers.”

A global luxury fashion brand uses Riskified’s AI-powered fraud and risk intelligence platform to confidently and quickly provide a premium experience to its high-value customer base while precisely identifying fraud and policy abuse.

The company reported:

• Approximately 75% reduction in chargebacks
• Conversion increased from ~50% to 75–80%
• Decrease in chargebacks spurred by rejected returns, from roughly $1M annually to $150K–$200K

The brand also improved its ability to identify repeat offenders across multiple identities, devices, and payment methods, enabling stronger protection against return abuse.

The report findings point to a new phase of ecommerce returns shaped by AI-driven fraud and abuse detection, identity-based intelligence, personalized return policies, and greater transparency around return costs. This shift reflects a growing gap between consumer expectations and merchant sustainability constraints, as behaviors many consumers see as normal are increasingly viewed by merchants as structural cost drivers. Merchants that succeed will be those able to distinguish between loyal customers and abusive behavior while maintaining seamless experiences for high-value shoppers.