Hacking Black Friday: using LLMs to save on the “sale of the year” | Kaspersky official blog
Source Kaspersky Security Blog
Description
When the sales hit, you might bag some serious bargains; however, you also have to watch out for unscrupulous vendors that just jack up prices. We’re bringing AI into the mix and suggesting working prompts designed to unlock genuine value.
Impact
Since the content does not describe a direct cybersecurity threat or vulnerability, the impact on European organizations is minimal from a technical security standpoint. The advisory helps consumers avoid financial loss and fraud during Black Friday sales, which can indirectly benefit organizations by reducing fraud-related chargebacks, reputational damage, and customer dissatisfaction. However, it does not pose a risk to organizational confidentiality, integrity, or availability. The mention of AI vulnerabilities is general and not tied to specific exploits. European consumers and businesses engaged in e-commerce could benefit from the guidance to detect and avoid scams, potentially reducing fraud losses. Retailers adhering to EU regulations like the Omnibus Directive may face increased scrutiny from informed consumers using AI tools. Overall, the impact is primarily on consumer protection and fraud prevention rather than on cybersecurity infrastructure or enterprise risk.
Defensive guidance
No direct mitigation is required for organizations as this is not a technical threat. For consumers and businesses involved in e-commerce, the following practical recommendations apply: 1) Use AI tools and LLMs to analyze price histories and detect suspicious price manipulations before purchasing. 2) Employ price-tracking services such as CamelCamelCamel, Keepa, AliPrice, and AliTools to gather objective pricing data. 3) Verify seller legitimacy by checking domain names, HTTPS status, and looking for signs of fake stores (e.g., unrealistic discounts, poor website quality). 4) Use AI prompts to vet offers and sellers critically, avoiding impulse purchases triggered by marketing tactics. 5) Maintain updated security solutions to protect against phishing and malware that may accompany scam attempts. 6) For organizations, educate customers about these risks and encourage transparency in pricing. 7) Monitor compliance with EU consumer protection laws to reduce deceptive marketing. These steps go beyond generic advice by integrating AI-assisted analysis and specific tools tailored for the Black Friday shopping context.
Summary
The Kaspersky blog post titled 'Hacking Black Friday: using LLMs to save on the “sale of the year”' is an advisory piece aimed at consumers preparing for Black Friday sales. It explains how retailers often manipulate prices by inflating them before the sales and then advertising large discounts that do not reflect real savings. The article advocates using AI-powered Large Language Models (LLMs) such as ChatGPT, Claude, and Gemini to analyze price histories, compare products, and detect fraudulent sellers. It provides step-by-step instructions on creating wish lists, tracking prices with third-party tools (e.g., CamelCamelCamel, Keepa), analyzing price dynamics to spot manipulations, searching for alternative products, and vetting sellers for legitimacy. The post also includes specialized AI prompts tailored to different LLMs to maximize their utility in shopping decisions. It warns about the increase in fake online stores during the Black Friday period and suggests indicators of scams, such as suspicious domain names and unrealistic discounts. While the article mentions that AI systems themselves have cybersecurity vulnerabilities, it primarily focuses on how consumers can use AI to protect themselves from financial scams and deceptive marketing tactics. No technical exploit, malware, or direct threat to organizational IT infrastructure is described. The content is educational and consumer-centric, promoting safer online shopping through AI assistance.