But yeah I had the same question: OP said that ChatGPT initially produced words that turned out to be sensible, but not where that later screenshot came from
> At the post office, a friendly staff member confirmed it could be sent, helped me package it up securely
It's actually at least half a case of not blindly trusting AI, and pressing until a positive answer is given:
> I asked ChatGPT how to send the laptop, and it gave me a spiel about finding a reliable freight service or courier.
> (Later) I should have listened to ChatGPT.