> ChatGPT got every single part of that wrong which chatgpt? and feeding the initial question to perplexity.ai yields a result that is practically identical t

> ChatGPT got every single part of that wrong which chatgpt? and feeding the initial question to perplexity.ai yields a result that is practically identical to the perfect answer. of course, the author has a point. as many here already stated it isn't even worth talking about it cause it is so obvious. then again as many correctly notice llms are sold with this promise. practically, the subject is way too niche - i'd assume - and probably there are also conflicting statements on the interweb which the model was trained on. having said that - perplexity got it right with web-access turned on.