3 ms·

> I doubt you actually believe Stephen King might be a p zombie and I'm skeptical you really think consciousness is an illusion. Consciousness is an unobservab

by roenxi 3y ago

> I doubt you actually believe Stephen King might be a p zombie and I'm skeptical you really think consciousness is an illusion. Consciousness is an unobservable, undefinable thing which with LLMs in the mix we can theorise has no impact on reality; since we can reproduce all the important parts with matrices and a few basic functions. You can doubt facts all you want, but that is a pretty ironclad position as far as logic, evidence and rationality goes. Consciousnesses is going the way of the dodo in terms of importance. > If I sawed off your leg would it have the moral consideration of removing the leg of a barbie doll's leg if you feel your consciousness is an illusion? For sake of argument, lets say conclusive proof arises that Stephan King is a philosophical zombie. Do you believe that suddenly you can murder him? No; that'd be stupid and immoral. Morality isn't predicated on consciousness. I'm perfectly happy to argue about morality but consciousness isn't a thing that makes sense outside of talking about someone being knocked unconscious as a descriptive state. > When I feel the desire to urinate, or post comments on hacker news, or admire a rainbow, or sleep am I also "doing statistics?" No, you're responding to stimulus. But right now it looks extremely likely that the creative process is driven by statistics as has been revealed by the latest and greatest in AI. Unless you can think of a different mechanism - I'm happy to be surprised by other ideas at the moment it is the only serious explanation I know of. > You seem to be referring to what people traditionally call "thinking" or "cognition" and rebranding it as "statistics" in search of some rhetorical point. I don't think I've said anything about thinking or cognition. Although statistics will crack those too, but I'm expecting them to be more stateful processes than the current generation of AI techniques. > Copyright existed far longer than the last 50 years so how is our 50 years of culture relevant? The U.S. has had copyright since 1790. Yeah but the law has been continuously strengthened since then and as it's scope increases the damage gets worse. The last 50 years are where new works are effectively not going to enter the public domain before everyone who was around when they were created is dead.

circlefavshape 3y ago
> Consciousness is an unobservable thing No it isn't. I observe consciousness in myself, right?
- roenxi 3y ago
  I doubt it. You don't have long enough to do so before time passes and you're relying on your memory being accurate. At which point it is more likely that you're something similar to an artificial neural network that has evolved to operate under a remembered delusion that you exist as an entity independent from the rest of existence. It is far too obvious why that'd be an evolutionary advantage to discount the theory. I'm not saying this has any meaningful implications for day to day existence, the illusion is a strong one. But LLMs have really shrunk the parts of the human experience that can't be explained by basic principles down to a wafer so thin it might not exist. In my view it is to thin to be worth believing in, but people will argue.
  - bamboozled 3y ago
    I'd really like to understand what you're trying to convey. If there is no direct experience, and it really is just an illusion, do you mind if I cut off your genitals ? Because even if existence is an illusion, being that illusion and experiencing it is still obviously relevant? Seems like a breathtakingly awkward position to take.
    - roenxi 3y ago
      Well I suppose two points: 1) Transexualism and congenital analgesia are both things, so that isn't a meaningful test of consciousness if that is what you are going for. 2) I personally would be extremely upset ... but anything going on with me being extremely upset can be explained with analogy to ChatGPT + a few hardware generations + maybe some minor tweaks that are easy to foresee but not yet researched. Everything that would happen would be built up from basic stimulus-response's mediated by a big neural net and a bit of memory. These ideas aren't that new, there have been schools of Buddhist thought that have been exploring them for around 2,500 years if you want a group that has thought through this more thoroughly. Probably other traditions that I haven't met yet. It is just that the latest LLMs have really put some practical we-know-because-we-have-done-it muscle behind what was already a strong theory.
      - 3y ago
staticman2 3y ago
Okay, first of all AI experts will tell you LLMs need a lot of data to do what they do, while humans do more with less data. I'm paraphrasing something Geoffrey Everest Hinton said in an interview. He also said current systems are like idiot savants. So why the "evidence" and "rationality" led you to say current systems are just like the brain, I don't know, seems a religious belief to me. Also most humans are not great at fiction writing, but just for the record GPT 4 is bad at long form fiction writing. It has no idea how to do it. Its training set is likely just small chunks of text and LLMs make up everything one token at a time, which isn't conductive to managing a 100k word or greater story. It's also been trained to wrap things up in 2000 tokens or less. It does this even if you tell it not to. It also isn't very creative, at least in contrast to a professional fiction writer. It also can't really follow the implications of the words it writes. So if it implies something about a character 300 words in it'll have forgotten that by 4000 words in because it doesn't actually have a mental model of the character. I mention this in case you are under the delusion Stephen King is going to lose his job to GPT 4, or that it actually is as creative as him. I mean, if a chatbot could write as well as king for 200,000 words in a coherent, original story with a beginning middle and end that would be fascinating and terrifying. But we're not anywhere close to there yet. >For sake of argument, lets say conclusive proof arises that Stephan King is a philosophical zombie. Do you believe that suddenly you can murder him? No; that'd be stupid and immoral. Morality isn't predicated on consciousness. Morality has a lot to do with consciousness. If I chop down a tree branch at my house, it isn't immoral because the tree is not conscious- it lacks a central nervous system so as far as we can tell there is no suffering. If I chop off your arm (or your gentiles as another poster suggested) your consciousness and the pain it suffers is what makes the action immoral. If Stephen King were a p zombie that would say a lot about the morality of destroying him. I don't even understand what belief system you are trying to communicate when you say things like consciousness is an illusion or has nothing to do with morality, or an LLM is the same thing as a brain. Like, I realize these are not your own ideas and you are repeating memes you've heard elsewhere, so it's not like you are speaking in tongues. But any human being presumably knows they have subjective experience (so why say consciousness is an illusion?) And anyone who has used ChatGPT know it doesn't operate like their brain (no sense of identity, agency, emotions, etc, no ability to form long term personal memories, no real opinion or preference on things) So I don't really get these takes.

by roenxi 3y ago

circlefavshape 3y ago

> Consciousness is an unobservable thing No it isn't. I observe consciousness in myself, right?

roenxi 3y ago
I doubt it. You don't have long enough to do so before time passes and you're relying on your memory being accurate. At which point it is more likely that you're something similar to an artificial neural network that has evolved to operate under a remembered delusion that you exist as an entity independent from the rest of existence. It is far too obvious why that'd be an evolutionary advantage to discount the theory. I'm not saying this has any meaningful implications for day to day existence, the illusion is a strong one. But LLMs have really shrunk the parts of the human experience that can't be explained by basic principles down to a wafer so thin it might not exist. In my view it is to thin to be worth believing in, but people will argue.
- bamboozled 3y ago
  I'd really like to understand what you're trying to convey. If there is no direct experience, and it really is just an illusion, do you mind if I cut off your genitals ? Because even if existence is an illusion, being that illusion and experiencing it is still obviously relevant? Seems like a breathtakingly awkward position to take.
  - roenxi 3y ago
    Well I suppose two points: 1) Transexualism and congenital analgesia are both things, so that isn't a meaningful test of consciousness if that is what you are going for. 2) I personally would be extremely upset ... but anything going on with me being extremely upset can be explained with analogy to ChatGPT + a few hardware generations + maybe some minor tweaks that are easy to foresee but not yet researched. Everything that would happen would be built up from basic stimulus-response's mediated by a big neural net and a bit of memory. These ideas aren't that new, there have been schools of Buddhist thought that have been exploring them for around 2,500 years if you want a group that has thought through this more thoroughly. Probably other traditions that I haven't met yet. It is just that the latest LLMs have really put some practical we-know-because-we-have-done-it muscle behind what was already a strong theory.
    - 3y ago

staticman2 3y ago

Okay, first of all AI experts will tell you LLMs need a lot of data to do what they do, while humans do more with less data. I'm paraphrasing something Geoffrey Everest Hinton said in an interview. He also said current systems are like idiot savants. So why the "evidence" and "rationality" led you to say current systems are just like the brain, I don't know, seems a religious belief to me. Also most humans are not great at fiction writing, but just for the record GPT 4 is bad at long form fiction writing. It has no idea how to do it. Its training set is likely just small chunks of text and LLMs make up everything one token at a time, which isn't conductive to managing a 100k word or greater story. It's also been trained to wrap things up in 2000 tokens or less. It does this even if you tell it not to. It also isn't very creative, at least in contrast to a professional fiction writer. It also can't really follow the implications of the words it writes. So if it implies something about a character 300 words in it'll have forgotten that by 4000 words in because it doesn't actually have a mental model of the character. I mention this in case you are under the delusion Stephen King is going to lose his job to GPT 4, or that it actually is as creative as him. I mean, if a chatbot could write as well as king for 200,000 words in a coherent, original story with a beginning middle and end that would be fascinating and terrifying. But we're not anywhere close to there yet. >For sake of argument, lets say conclusive proof arises that Stephan King is a philosophical zombie. Do you believe that suddenly you can murder him? No; that'd be stupid and immoral. Morality isn't predicated on consciousness. Morality has a lot to do with consciousness. If I chop down a tree branch at my house, it isn't immoral because the tree is not conscious- it lacks a central nervous system so as far as we can tell there is no suffering. If I chop off your arm (or your gentiles as another poster suggested) your consciousness and the pain it suffers is what makes the action immoral. If Stephen King were a p zombie that would say a lot about the morality of destroying him. I don't even understand what belief system you are trying to communicate when you say things like consciousness is an illusion or has nothing to do with morality, or an LLM is the same thing as a brain. Like, I realize these are not your own ideas and you are repeating memes you've heard elsewhere, so it's not like you are speaking in tongues. But any human being presumably knows they have subjective experience (so why say consciousness is an illusion?) And anyone who has used ChatGPT know it doesn't operate like their brain (no sense of identity, agency, emotions, etc, no ability to form long term personal memories, no real opinion or preference on things) So I don't really get these takes.