Walvis Eylant

My prompt, and ChatGPT's response. That it can more or less comply with a request is certainly remarkable. I'd have no idea how to create such software; I only know a little BASIC, html, css, and wiki.  That said, there's any number of issues and weirdly they would seem to have been easily avoidable. It doesn't give the poem a title, even though they're pretty standard for poetry; that's something I've noticed it didn't tend to do in the past. The poetry previously mentioned when it had forged references; in that case it did. Why would it describe Tommo and Toby as children? Why would it force a rhyme by changing the word to the singular, instead of looking for a different rhyme? Granted, poetry often takes liberties with words - maybe that's what it's trying to do? Why does it describe Tommo as being "with tattoos" when he has none, and gets none, and has a horror of getting them? Isn't it a little strange to describe Fayaway as &q

Search engine capability seems to be a big part of what the creators of ChatGPT currently envision it doing, regardless of how it was originally conceived.  After login, one of the three columns of text on the screen above the user’s prompt window has:                                     Examples “Explain quantum computing in simple terms” “Got any creative ideas for a 10 year old’s birthday” “How do I make an HTTP request in Javascript" To my mind that main opening screen page bears more resemblance to Google’s main search page than to a chat window, something that might influence users' expectations and interactions. Regardless, it should be able to have substantive conversations; there’s enough human BSers that we don’t need artificial ones. Under the third column of text, Limitations, it has: May occasionally generate incorrect information May occasionally produce harmful instructions or biased content Limited knowledge of world and events after 2021 That first disclaimer