SiliconAngle reports that a Netcraft study has revealed an increase in the use of artificial intelligence large language models in crafting fake online stores and deceptive content during the Black Friday shopping period.
Between August and October, the detection of fraudulent e-commerce sites rose by 110%, with tens of thousands of these hosted on SHOPYY, a Chinese e-commerce platform exploited by cybercriminals. Over 9,000 new fake store domains were identified on SHOPYY alone between November 18 and 21.
Large language models play a central role by generating convincing product descriptions, often rephrased from legitimate sites like Amazon. This tactic not only mimics trusted retailers but also boosts visibility in search engines through optimized keywords. Cybercriminals also use these tools to rewrite titles and descriptions to bypass plagiarism detection and create unique, polished listings. The efficiency of large language models allows rapid scaling of fraudulent operations, enabling cybercriminals to manage extensive campaigns with minimal effort.
These activities thrive during heightened shopping seasons like Black Friday and Cyber Monday, targeting unwary consumers "To combat the fake store threat, retailers must invest in proactive security measures, educate customers, and actively monitor for fraudulent activity and brand impersonation," the report said.
