Ecommerce fraud surges as AI-driven attacks grow

Ecommerce fraud rose 33% in the first four months of 2026, fueled by artificial intelligence and organized crime, a new industry report found.

The findings, based on transactions from thousands of online merchants and 950 million digital wallets, show fraud now spans the entire customer journey, from account creation to returns.

Account takeovers and card-testing lead the increase

Account takeover attacks climbed 78% year over year, while card-testing incidents—where criminals test stolen card details—surged 175%. Fraud linked to “buy online, pick up in store” options grew 65%, and first-party fraud, involving customers exploiting return policies or promotions, increased 9%.

The fraud prevention firm Signifyd compiled the data. Its analysis shows AI tools have reduced the cost and complexity of attacks, enabling fraudsters to expand operations quickly. Organized crime groups and rising consumer abuse are also changing how merchants evaluate risk.

Raj Ramanand, Signifyd’s co-founder and CEO, said retailers can no longer view fraud as a problem limited to checkout. “Ecommerce is entering a new phase where AI is accelerating both innovation and fraud,” he said. “Protecting revenue and preserving the customer experience are now the same challenge.”

Fraud tactics become more advanced

The report details a shift in fraud economics, with AI enabling larger, faster attacks at lower cost. Examples include stolen identities, fake websites harvesting personal data, and schemes tied to high-value items like iPhones, often used for money laundering.

Nicole Jass, Signifyd’s senior vice president of enterprise strategy, said attackers combine multiple tactics to avoid detection. “What’s changing isn’t just the volume of fraud, but how it works,” she said. “The line between organized fraud and consumer abuse is fading, creating a tougher environment for retailers.”

Fraud’s effects now go beyond chargebacks. Account takeovers can expose customer profiles, while false returns and fulfillment abuse disrupt inventory and store operations. Retailers must balance tighter fraud controls with a smooth experience for legitimate shoppers.

This challenge grows as AI-assisted shopping expands. The report describes an emerging “autonomous commerce” environment, where transactions are increasingly driven by software agents. Merchants now need to verify trust in both human and machine-driven interactions.

Digital identity has become a key target. Criminals exploit weak authentication systems, forcing retailers to rethink customer verification without adding unnecessary steps.

Fraud prevention once focused on payment authorization. Risk assessment now covers account activity, order fulfillment, and returns. The change reflects broader shifts in retail, where customer experience and security overlap.

The findings show fraud is no longer just a payment team issue. It affects operations, customer service, and marketing, as false positives—legitimate orders flagged as fraud—can drive shoppers away.

Signifyd’s data also highlights a geographic split. Fraud pressure is highest in markets with rapid ecommerce growth, such as Southeast Asia and Latin America, where digital payment adoption outpaces fraud prevention tools. In the U.S. and Europe, attacks are more targeted, often focusing on high-value items or weak return policies.

Retailers are responding with technology and process changes. Some use AI-driven fraud detection, while others tighten return policies or require extra verification for high-risk orders. The difficulty, Jass noted, lies in doing so without harming the shopping experience.

She added that connecting patterns across merchants will help address the problem. “It’s not just about stopping fraud—it’s about maintaining trust in a system where both humans and software agents make decisions.”