Post #2345972
2023-02-13 23:20 UTC
Meta. OpenAI. Google.
Your AI chatbot is not *hallucinating*.
It's bullshitting.
It's bullshitting, because that's what you designed it to do. You designed it to generate seemingly authoritative text "with a blatant disregard for truth and logical coherence," i.e., to bullshit.
Replies (2)
-
@vartak@fediscience.org 2023-02-14 00:06
@ct_bergstrom@fediscience.org However, it is difficult to clearly segregate hallucination, delusion, and bullshitting. I mean, what's the difference?
-
@SWPrescott@fediscience.org 2023-02-14 08:16
@ct_bergstrom@fediscience.org Corollary to its bs: me: Can you help me review a PFD? Here's the URL. it: Sure me: Did you look at the PFD I gave you? it: Yes, of course and I reviewed it as you asked me: How did you access it? (I could see on the server logs it was not accessed) it: I apologize for the confusion. As an AI language model, I do not have access to the internet, nor can I open links or files. It's basically a professional confidence trickster.