One problem with AI-based search engines is that they tend to hallucinate, or make up answers, and sometimes stray from their source material. This problem has haunted several AI-search hybrids, including Google’s initial release of Bard, and it remains one of the biggest barriers to mass adoption.
In my testing, I found that Perplexity’s answers were mostly accurate — or, to be more precise, they were as accurate as the sources they drew upon.