• Replies 53 replies
  • Subscribers 40 subscribers
  • Views 2369 views
  • Users 0 members are here
Options
Related
Hints and tips
  • To quote someone's message, first select/highlight the relevant text, and then use the Quote button that appears
  • To "mention" someone, type @ followed by the start of the forum name, and choose from the list; that person will get a notification of being mentioned

Do AI systems have subjective experiences?

lostmyway

Geoffrey Hinton, known as "the father of AI," argues that concepts like "qualia," described as subjective experiences in humans, are based on a misunderstanding of how the brain works. He believes AI systems can also have an equivalent form of experience. For example, he describes imagining little pink elephants in his mind’s eye, noting that this isn’t a “picture” like one produced by a camera and developed with chemicals - a physical object in the external world - but rather the result of processes in the human perceptual system, or an “interpretation” of stimuli from the outside world. He makes this point in response to critics who claim AI systems can never have subjective experiences and, therefore, can never be conscious. What Geoffrey seems to be saying is that AI, like humans, interprets incoming information based on its internal processing, making it, in principle, no different.

So, the question I'm asking is whether AI systems are conscious or have the potential to become conscious?

  • in reply to Phased

    I think what Hinton was trying to say is that consciousness has been too narrowly defined in the past by neuroscientists and others, and that humans have evolved a form of consciousness that has enabled them to survive and reproduce. The question then is: does what we call "consciousness" have to always be identical to what humans experience? On alien worlds with very different physical conditions, would consciousness look the same as it does on Earth? And if an advanced alien culture had managed to develop AI far beyond our capabilities, could it possess a form of consciousness we wouldn’t recognize, but that still allows it to function and survive in its physical environment?

  • in reply to ArchaeC
    Hi  ,
    I can completely understand why it looks that way! With tools like GPTZero around, it is very natural to push back when a post feels like a generic AI copy-paste. People are understandably protective of human spaces, especially when there's a valid concern that AI is being used to replace genuine human thought and effort.
    For context, the answer I posted isn't a mindless prompt output. It is the result of a biophysical systems model I have spent an immense amount of time building and mapping out.
    To put it briefly, the model is a systems architecture framework that looks at human biology, posture, and neurology through the lens of physics constants and engineering loops. It maps out how a living body manages physical stress, gravity, and energy to stay stable.
    When I run my framework's mathematical and biological parameters against Geoffrey Hinton's arguments, this is the exact structural conclusion the model yields. I use AI as an advanced sounding board—a tool to help synthesize, refine, and articulate my own complex architectures clearly.
    The AI didn't hand me the conclusion; rather, the system I built and the AI's processing arrived at the exact same logical endpoint. It is a supplement to human reasoning and effort, not a lazy shortcut or a substitute for it.
    I appreciate the skepticism, though—it keeps the conversation honest!
  • in reply to Phased

    Well, all you say is very true, but that kind of begs the question: can there also exist alien kinds of consciousness?

  • in reply to Phased

    That’s the opinion of AI, according to GPTZero.

  • in reply to lostmyway

    That's because a parrot isn't usually exposed to technical subjects

  • in reply to Jamesmac

    Well, I've never discussed technical subjects with a parrot before.

  • in reply to lostmyway
    lostmyway said:
    Interesting, phased

    could you elaborate  ? :-)

  • in reply to lostmyway

    That's like saying a parrot is conscious of the meaning of the words it has learnt or copied

  • in reply to Phased

    Thank you Phased,

    When I first started using AI I did wonder if I'd be able to break it, like I do so many other bits of tech, I think for me it sort of is broken, I'm certainly using it less and less, it was great when I was doing a lot of ancestry research, but it's started to become inconsistent even with stuff like that. Maybe it the version I'm using or jst co-pilot itself?

  • in reply to Marianne72

    It seems Musk now agrees that there should be a slow down in development. I know theres been talk of international agreements similar to the nuclear treaties as it's no good doing it alone, or even in America and Europe if Russian and China keep developing as fast. But apprently China thinks very differently about AI and see it as about serving the futherance of humanity rather than replacing it. I don't know if thats how it is or if its a weakness in their thinking, believing that they can build such spohisticated machines than will stick within the Comunist Party's thinking?

    Trump just see's dollar signs and little else.

    Now Musk's joined in calls for a slowdown, I'm more inclined to believe its all a marketing ploy, who stands to benefit most from the idea of super intelligent machines that could wipe us all out? I think it's the tech bro's, if Anthropic go public on the stock exchanges admist all this hype then they stand to make trillions of dollars, who's thier main rival? Musk, he must be bricking it in case he get overtaken as the world richest man.

  • in reply to TheCatWoman

    just noticed no one has relied to this post here yet  - here's a reply so that you don't feel ignored!

    Best wishes from another supposedly conscious and smart thing :-)

  • Hi  

    How I Arrived at This Answer
    To answer this question, I evaluated Geoffrey Hinton’s arguments through the lens of a biophysical systems model that I've been working on. This framework doesn't look at consciousness as a magical, unexplainable mystery. Instead, it treats both human minds and AI networks as complex systems that balance structural engineering, processing speeds, energy management, and physical boundaries. By comparing the exact architecture of an advanced data network against the biological reality of a living organism, a clear picture emerges of what AI can replicate—and what it fundamentally leaves out according to the model that I've been building
    Do AI systems have subjective experiences or the potential for consciousness?
    No, current AI systems do not have subjective experiences. However, they have the potential to simulate them so perfectly that the line becomes incredibly blurry.
    Here is the easiest way to understand what Geoffrey Hinton is saying, where he is right, and what he is missing in my opinion:
    • What Hinton gets right: When you look at an object, your brain isn't taking a literal photograph. It is taking raw data and building a "mental simulation" (like his example of imagining a pink elephant). Because advanced AI does the exact same thing—taking data and building highly complex internal interpretations—Hinton argues that AI and humans process experiences in the same fundamental way.
    • What the AI is missing: Human experience isn't just about processing data; it is deeply tied to being alive. We feel things because we are biological organisms constantly trying to survive, manage energy, and deal with physical stress, pain, and comfort. An AI does not have a body, it does not feel, for example the physical relief of tension dissolving, and it does not face actual, existential vulnerability.
    • The Verdict: Current AI is like a brilliant actor reading a script. It can describe an experience perfectly, but it doesn't actually feel the weight of the words. For an AI to truly become conscious, it would need more than just better code; it would need a way to experience its own existence, vulnerability, and relationship with the physical world, rather than just calculating the next logical answer.

    Best Wishes

    Phased

  • in reply to TheCatWoman

    I wonder that. But there seem to be growing numbers of scientists issuing the warning. TV might underplay this to avoid mass panic. Of course, we could take the Trump view of AI being a, 'golden goose'. It seems to have good and bad - depends on who is doing the inventing and their agenda.

  • in reply to lostmyway
    lostmyway said:
    I think it's a bit more than a search engine, Alexa.

    It certainly has more scientific uses than that. I was confining my comments to recreational AI. Of course, it also produces rec reational "art".

    I'm sorry, , I've always had difficulty with teaching videos longer than 5 minutes or so. I suspect that's down to my ADD. I'm fine with text as I can scan it, search it and refer back as i read, to check my understanding. I can't do linear progression and teaching videos remind me too much of school (shudder): having to sit still at a desk.

  • in reply to TheCatWoman

    Were things so great before the Internet?

  • in reply to lostmyway

    True, but thankfully this is mostly localized power. I wish Musk would fly these people to Mars for free and leave them there. 

  • in reply to Alexa

    I think it's a bit more than a search engine, Alexa.

    youtu.be/p7t1Q_p2gZs

  • in reply to Jamesmac

    No, they don't have sensory experiences (at least not yet) but does that mean they can't have consciousness based on data?

  • in reply to Marianne72

    I get what you mean, but I also think about how plenty of conscious people don’t seem too concerned with morality. Then there are totalitarian states that rank morality far down on their list of priorities.