Perplexity, an AI search engine touted as a game-changer, is under fire for citing AI-generated blog posts as authoritative sources. Despite CEO Aravind Srinivas’ claim that “citations are our currency,” the startup’s search engine has been found to draw information from and cite AI-generated posts on various topics, including travel, sports, food, technology, and politics. A study by GPTZero, an AI content detection platform, revealed that Perplexity’s search engine is prone to citing inaccurate, outdated, and contradictory information from AI-generated sources.
The study, which analyzed over 100 prompts, found that Perplexity users only need to enter three prompts before they encounter an AI-generated source. In some cases, Perplexity’s responses included out-of-date information, and the startup’s system even cited an AI-generated LinkedIn post as a reference. The reliance on low-quality web sources raises concerns about the spread of misinformation and the potential for “second-hand hallucinations.”
Perplexity’s Chief Business Officer, Dmitri Shevelenko, acknowledged that the system is “not flawless” and that the company continuously improves its search engine by refining its processes to identify relevant and high-quality sources. However, experts argue that the company’s algorithms are not perfect and need to be continually refined, especially as AI-generated content becomes more sophisticated.
