I was using ChatGPT to assist me with using Ghidra [0] in a recent CTF I was participating in. I wanted to see if I could find answers quicker with a chatbot ov

I was using ChatGPT to assist me with using Ghidra [0] in a recent CTF I was participating in. I wanted to see if I could find answers quicker with a chatbot over looking up videos, tutorials, etc. through the normal methods. The thing that really surprised me about ChatGPT wasn't in the details of responses it gave, or how slow it was, or anything like that. Rather, it was how much it appeared to be making up out of thin air. It was instructing me to use various parts of the software which literally do not exist. It was referencing features which are not there, and attempting to "correct me" when I said I couldn't use those instructions. All in all, what would have taken about 5-10 minutes of searching took about 25 minutes, with no resolution and me having to scold the chatbot for repeatedly giving instructions that were flat out wrong. [1] On a very, very superficial level, when you just need basic instructions on, say, how to tie a tie or whatever, it seems "amazing" because of how authoritative it sounds. I have noticed though that when you start to get into even some of the basic levels of instruction or details on a given topic, it seems to defer to the superficial responses first, then resort to making things up, giving outright incorrect information, or giving extremely vague responses which can't really be taken one way or another. To give credit where credit is due, it's an achievement that you can feed it a question on just about any topic and get some kind of coherent response. It's just a shame that, for all of the hype and marketing around it, once you take any sort of a closer look at the product itself, you very quickly start to see cracks when it comes to details. [0] https://en.wikipedia.org/wiki/Ghidra https://en.wikipedia.org/wiki/Ghidra [1] https://pastebin.com/42juVPfM https://pastebin.com/42juVPfM