When Recalls Go Rogue: The Impact of AI-Fueled Recall Fraud

Recall fraud[1] is a growing and increasingly sophisticated problem that occurs when bad actors exploit online recall processes to obtain remedies to which they are not entitled. This illicit activity increases compliance costs for companies, undermines the effectiveness of legitimate recall programs, and results in heightened requirements to submit a legitimate request for a recall remedy.

On April 15, 2026, the U.S. Consumer Product Safety Commission (CPSC) issued a Request for Information (RFI) on recall fraud and its impact, seeking information from manufacturers, importers, distributors, retailers, recall administrators, consumer advocates, and the public. The CPSC’s RFI sought input on the following four issues: 1) observed recall fraud; 2) impacts of recall fraud; 3) fraud mitigation measures, and 4) potential Commission actions. The RFI specifically calls out Artificial intelligence (AI)-generated fake submissions and coordinated redemption schemes as key concerns. The CPSC also invited feedback on existing fraud detection tools and technologies used to address evolving organized fraud schemes, including AI capabilities.

During the RFI comment period, which ended on June 15, 2026, the CPSC received twenty comments from manufacturers, industry organizations, private individuals, and recall administration and logistics companies that provide support to recalling companies. A few central themes emerged in these comments, including: 1) concerns over AI-fueled fraud and the shift to authenticated capture, 2) product return as the safest and most effective way to prevent recall fraud, and 3) the costs of recall fraud.

1. AI-Fueled Fraud and the Shift to Authenticated Capture

Generative AI emerged as the most frequently referenced topic among recall administrators and technology vendors involved in recall logistics. Generative AI has fundamentally transformed the recall fraud landscape by making it easier to fabricate images, receipts, photographs, serial numbers, and identity documents. More troubling, industry participants noted that generative AI has increased both the sophistication and scale of recall fraud, with trends suggesting that fraud now operates on an organized, systemic level rather than as isolated consumer opportunism.

For example:

• Sedgwick, an industry leader in product recall management, noted that the rise in AI-generated fabricated images has rendered photo-based verification unreliable. Sedgwick emphasized the need for fraud mitigation during the “entire recall lifecycle” and the use of appropriate technology and controls that balance “convenience with security.”
• Truepic, a software platform that enables consumers to verify product authenticity through real-time photo capture, stressed the importance of taking a live photo, as opposed to uploading a previously taken photo, described as “live capture” functionality to guard against AI-generated images.
• Bawte, Inc., a consumer engagement platform that “operates a QR-code-based product registration and post-purchase engagement platform for consumer product brands,” similarly addressed AI-fueled fraud, noting that uploaded receipt photos can no longer reliably authenticate purchases given how easily consumers can generate fabricated images.
• The National Association of Manufacturers (NAM) reported similar findings, observing that AI-generated images now feature “convincing faulty material, handwritten markings, and hyper-company-specific product errors.” NAM further urges the CPSC to implement advanced image verification technologies to detect AI-generated photos, “photo of photo” submissions, and duplicated or manipulated photos.

It seems that most industry participants agree on a shift toward fraud detection at the point of capture, assessing authenticity based on a photograph taken in real time along with its metadata (such as geolocation and image characteristics) instead of post-submission verification of eligibility for the recall. These comments reflect a broader understanding that traditional recall verification methods are no longer sufficient in an AI-enabled environment, and that proactive, risk-based detection must become standard practice.

2. Returning the Product as the Safest and Most Effective Way to Prevent Recall Fraud

Although technology and software exist to authenticate and verify recalled products from consumers’ homes, several commenters emphasized that physically returning the product to the recalling company is still the safest and most effective way to prevent recall fraud.

Recall Results, a recall management partner, emphasized this point, stating that “[t]he number one way to prevent recall fraud and abuse has historically been, and will continue to be, the actual return of the recalled product to the recalling company or to their recall administrator for destruction.” An anonymous comment reiterated that recalling companies should physically obtain the recalled product—either by mail or in-person drop-off facilities—as this will not only prevent fraud but bolster consumer safety. The Toy Association maintained this position in their comment as well, stating that the most effective way to prevent recall fraud is to require the return of the product or the smallest critical component to the recalling company. Sedgwick commented similarly, recommending “physical returns where feasible.” Product return also reduces the risk that a recalled product will be re-sold by the consumer, even after receiving a refund—an issue that the CPSC is particularly concerned about. 

However, this does not come without burden to consumers, as noted by several individual commenters. If consumers are required to return products to receive a remedy, consumers must do more than simply fill out a form. Additionally, consumers may lose further confidence in the brand when they feel they must do more work to solve a recall issue they never asked for. This additional burden is also likely to reduce consumer participation in recalls, potentially resulting in the continued use of recalled products.

3. The Costs of Recall Fraud

Recall fraud has a financial impact on all participants in the process but is most acutely felt by the recalling company who is already paying a considerable price. According to NAM, a product recall has direct and indirect costs, the reported numbers for which represent “less than half the total amount spent.” As reported, “[d]irect recall operations account for approximately 35% of costs—around $10 million—while indirect costs from business interruption and rehabilitation drive final costs into the $30-50 million range.”

Recall fraud only adds cost and complication. Sedgwick cited an example of “20,000 suspected fraudulent claims, representing a potential multimillion dollar financial exposure” that would historically have been paid and included in regulatory reporting of corrections, “potentially misrepresenting true consumer engagement.” Efforts to mitigate and prevent fraud are also expensive. As Sedgwick noted, there are many compliance and operational costs, including manual claims review and investigation when automated systems flag suspicious submissions, resources to identify fraudulent trends, increased customer service volume, returns processing for non-recalled or junk items, and legal and compliance support costs. Scammers are also filing complaints with the CPSC and other agencies in an effort “to pressure companies into paying questionable claims,” placing “a significant burden on internal resources, requiring time and coordination across customer service, legal, and compliance teams.”

Key Takeaways

Fraudsters are using generative AI to fabricate serial numbers, receipts, and images to obtain refunds or replacements for products they never bought, often as part of broader coordinated schemes.  At present, authenticated capture is the most effective method to combat these efforts.

While old school fraud prevention methods like physical product returns are the most effective at preventing fraud and removing recalled products from the field, they also require the most effort by the consumer, causing further loss of brand confidence and reducing recall participation. 

Fraud increases costs for everyone involved in a recall, but recalling companies are most directly impacted. Ultimately, stakeholders must work together to find a solution to this evolving problem.

Thank you to Summer Associate Christina Hanssen for her contributions to this article.

[1] Recall fraud happens when bad actors trick companies out of refunds, replacements, or repairs for recalled products. These scammers use fake receipts, alter photos of products, or use artificial intelligence to generate fake images of destroyed goods to collect money they do not deserve. See also Consumer Product Safety Commission, Request for Information: Consumer Product Recall Fraud, Regulations.gov (Apr. 15, 2026), https://www.regulations.gov/document/CPSC-2026-0067-0001 (“the term ‘recall fraud’ refers to fraudulent or abusive conduct associated with the redemption of recall remedies, including the submission of false or misleading information, fraudulent redemptions, or other schemes to abuse recall programs”).