ChatCPT - Creating the answers.

I asked ChatGPT "Is ChatGPT using an established set of answers to preconceived list of questions, or does it create the answers from scratch?" It's a decent question. The AI on those 'chat with us' pages appear to only have a few set questions they can answer with the only other response being 'I don't understand, contact human' (I'm paraphrasing here, and they can seem to identify the question even if you phrase it a slightly different way).

Anyway, so I'm wondering if ChatGPT is smarter than other AI simply because it has a bigger list of answers to read from, if it's copying from other sources on the internet or if it is actually writing responses to our questions.

Here's what it said:

ChatGPT does not use an established set of answers to preconceived questions. Instead, it generates responses based on patterns it has learned from the vast amount of text it has been trained on. The training data for ChatGPT consists of a wide range of sources, including books, websites, articles, and other text documents.

O-O I wonder how much text it had to read to get this smart. ChatGPT don't actually tell me the answer, I presume it will be some obscene amount. I don't actually want to know.

When you ask a question or provide a prompt to ChatGPT, it processes the input and generates a response based on the patterns and information it has learned during training. It does not retrieve pre-existing answers but rather generates a new response on the fly. This means that the answers it provides are not predetermined or fixed but are generated based on the context and information it has learned from its training data.

Nice, nice. The pattern thing is a little alarming, but we humans are creatures of habit. Considering it is generating new responses on the fly it would be fantastic at making on the spot speeches. I wonder if in future ChatGPT will be writing the politician's response to live media questions.

It's worth noting that while ChatGPT can provide helpful and informative responses, it's important to critically evaluate and verify the information it provides. As an AI language model, ChatGPT's responses are based on statistical patterns in the data it was trained on and may not always be accurate or up to date.

Sounds reasonable. I've only done a few questions on ChatGPT so far (considering as starting this blog was the first time I had used ChatGPT) but I kinda like how it keeps reiterating the importance to evaluate its response as it may not always be correct.