GoKawiilStudy: ChatGPT, Gemini and Perplexity cite fake retail sites in shopping answers
IP protection firm Podqi tested ChatGPT, Gemini, and Perplexity with shopping questions about 151 brands, generating 6,795 answers, and found all three chatbots sometimes surfaced risky websites such as impersonator sites, knockoff sellers, reregistered old domains, or inactive dead sites. Perplexity showed risky sites in 15% of answers, ChatGPT in 8%, and Gemini in 4%, with ChatGPT warning users about risky links it surfaced 24% of the time.
GoKawiil's interpretation of the reporting above, not reported fact.
As more shoppers turn to AI chatbots instead of search engines to find where to buy products, these findings suggest such tools could inadvertently funnel users toward scam sites or counterfeit sellers, particularly for smaller brands with less online visibility. The variation in warning behavior across chatbots indicates safety safeguards remain inconsistent, which could pressure AI companies to improve source verification as commerce-related queries grow.
- Podqi's study found all three major AI chatbots recommended risky shopping sites in some answers.
- Perplexity had the highest rate of risky site citations at 15%, compared to 8% for ChatGPT and 4% for Gemini.
- ChatGPT was most likely to warn users about risky sites it surfaced, doing so in 24% of such cases.
Source: theverge.com — Mia Sato, 2026-11-12
Published there as: “AI chatbots are recommending fake retail websites, study finds”
Read the original report → The summary and analysis above are GoKawiil's own, written from reporting by the source above. Facts and quotes belong to the original publisher.