Recent research warns that AI chatbots may actively recommend counterfeit shopping websites. Tech media outlet The Verge reported on an investigation by Podqi this autumn, highlighting a typical case: when a user asked Perplexity for the official website of the streetwear brand Stüssy, the AI directed them to a counterfeit page, presenting it as the genuine official site and claiming that authentic products could be purchased there. Among the three models examined, ChatGPT is relatively safer: when it provides risky websites, 24% of the time it includes a risk warning.
