I run a word-search API and I now regularly get emails from frustrated users who complain that it doesn't work the way ChatGPT tells them it works. For example

I run a word-search API and I now regularly get emails from frustrated users who complain that it doesn't work the way ChatGPT tells them it works. For example, today someone asked me why a certain request fails, and it turned out to be a fake but plausible URL to my API that ChatGPT had invented in response to "Does the Datamuse API work in French?" (It does not, and there's no indication that it does in the documentation.) Adding up all the cases like mine out there -- the scale of the misunderstanding caused, and amount time wasted, must be colossal. What bothers me is that not only has OpenAI extracted and re-sold all of the value of the Web without providing any source attribution in return; but they do so lyingly a good chunk of the time, with someone else bearing the costs.