According to WIRED, when the publication tested ChatGPT’s ability to accurately report on their own product reviews, the AI provided incorrect information. The test specifically asked ChatGPT to identify which TVs, headphones, and laptops WIRED’s reviewers had actually tested and recommended as the best options.
WIRED reports that ChatGPT’s answers were “all wrong,” highlighting a significant gap between what the AI claimed about the publication’s recommendations and what WIRED’s reviewers had actually tested and selected. The experiment appears designed to test whether ChatGPT could reliably surface accurate information from specific, verifiable sources—in this case, WIRED’s own published reviews.
The findings underscore ongoing concerns about AI chatbots providing confident-sounding responses that don’t align with factual information. According to WIRED’s reporting, users seeking product recommendations based on what the publication’s reviewers have actually endorsed would receive misleading guidance from ChatGPT. The story suggests that relying on AI chatbots for such specific, fact-based queries may lead to inaccurate results, even when the information being requested is publicly available and verifiable.