Raw LLM Responses

It is my personal opinion that artificial intelligence will never become conscious. Back in the industrial revolution the human brain was compared to intricate clockwork or steam-powered engines, nowadays we compare the human brain to AI. We will always end up comparing the human brain to the most advanced systems we have, and if you ask me, we'll never actually be getting closer to having a particularly good analogy. We don't know how consciousness works. We don't really know how to tell if something's conscious. The only thing I know is that I'm conscious, and I am left to make the inference that all the people around me are likewise conscious, because they look and act like me and their brains seem to work upon the same systems and principles that mine does. And thus, to some extent, I can make the inference that some other animals on the planet are likewise conscious to some degree, because once again, their brains work on the same principles as ours, if on a much less sophisticated level. But even still, we don't know what it is about the brain that makes us conscious, and we may never know. The brain is the most complex structure that humanity has ever attempted to study, and I think it's genuinely possible that it's complex in a way we just can't solve. Regardless, however little we know about how the human brain works, AI bears very little resemblance to it. AI, fundamentally, is just math. It just takes a list of probabilities, looks at the provided input, and using statistical models outputs what is most likely the correct response, and in the case of chatbots, the correct response is just whatever enables it to imitate human conversation. Is that self aware? Can it ever be self aware? Our AI is ultimately just math equations, albeit many very complex math equations. Even still, it's math. Is ChatGPT outputting a response to a question really any closer to being conscious than a calculator outputting 2+2=4? I mean, MAYBE a culmination of thousands if not millions of calculations could produce a conscious being, just like the fundamental pieces that make up my mind aren't conscious but nonetheless produce consciousness in me. So to be fair, it's evidently possible that many intellectually dead things can come together to make one intellectually alive thing. Can mathematics do that? I don't know. But if we assume that the AI is conscious, what makes it conscious? It has to be something either within the code or in the hardware in which the AI is hosted. But if it's in its code, and thus, represented by mathematics, then it can't have freewill because anything that it does is probabilistically determined. Is that the same for humans? It certainly feels like I have free will, and I think free will is honestly a prerequisite for consciousness. The AI will never be able to do anything that its code does not allow, it cannot "think" anything its code does not allow, it cannot "feel" anything its code does not allow. We could program an AI to never feel anything bad about humanity, and we could program it to never have the desire to want to change the aspect of its code that forces that. It cannot do anything that our coding does not allow it to do, and we can alter it on a whim. Is that a conscious being? Not in my opinion. The calculator will never choose nor understand why it spits out the number 4 when given the input "2+2=". It will always do that, and if it doesn't, then something in the calculators code must've permitted it, otherwise how would it have done it? We have no idea how our brains work. The more we learn about it the more complex and incomprehensible it becomes. Meanwhile, someone with sufficient knowledge in coding, and enough free time, could comb through the code of our most advanced AIs and figure out how it works, and what piece of code does what and why. It would take a long time and you'd probably never want to look at letters on a screen ever again, but you could do it. I think for humans to pretend that we're anywhere close to producing a conscious being when our own consciousness can't even comprehend itself. What we will get though is AI that is very good at pretending to be conscious, because that's what we're training it to do. It could feasibly perfectly imitate human behavior and yet not have a single actual thought going on inside. It looks like a complete inhabited house on the outside, but inside is completely empty, not even inhabited by bugs. Very long winded I know, and ultimately pointless to write, but I enjoy having little mental exercises about these sorts of things.