[DSMan195276]

The problem is the messy in-between, plenty of people who talk to professionals or call hotlines don't know that their questions are dumb. The bot should at a minimum say "I have no information on that" or "that's not a good idea", it should definitely not start giving nonsense recommendations just to reaffirm the question.

In other words you'd be pretty surprised if a real person in this context gave an answer even remotely close to what this chat bot gave. You can't expect a general person to know when the chat bot isn't giving back good information just because they asked something outside the norm.