Re: How would a computer know if it were conscious?
At the very least could it be said the AI is conscious of the question? Would this awareness of even a single piece of information be sufficient to make it conscious? Jason On 6/2/07, Hal Finney [EMAIL PROTECTED] wrote: Various projects exist today aiming at building a true Artificial Intelligence. Sometimes these researchers use the term AGI, Artificial General Intelligence, to distinguish their projects from mainstream AI which tends to focus on specific tasks. A conference on such projects will be held next year, agi-08.org. Suppose one of these projects achieves one of the milestone goals of such efforts; their AI becomes able to educate itself by reading books and reference material, rather than having to have facts put in by the developers. Perhaps it requires some help with this, and various questions and ambiguities need to be answered by humans, but still this is a huge advancement as the AI can now in principle learn almost any field. Keep in mind that this AI is far from passing the Turing test; it is able to absorb and digest material and then answer questions or perhaps even engage in a dialog about it. But its complexity is, we will suppose, substantially less than the human brain. Now at some point the AI reads about the philosophy of mind, and the question is put to it: are you conscious? How might an AI program go about answering a question like this? What kind of reasoning would be applicable? In principle, how would you expect a well-designed AI to decide if it is conscious? And then, how or why is the reasoning different if a human rather than an AI is answering them? Clearly the AI has to start with the definition. It needs to know what consciousness is, what the word means, in order to decide if it applies. Unfortunately such definitions usually amount to either a list of synonyms for consciousness, or use the common human biological heritage as a reference. From the Wikipedia: Consciousness is a quality of the mind generally regarded to comprise qualities such as subjectivity, self-awareness, sentience, sapience, and the ability to perceive the relationship between oneself and one's environment. Here we have four synonyms and one relational description which would arguably apply to any computer system that has environmental sensors, unless perceive is also merely another synonym for conscious perception. It looks to me like AIs, even ones much more sophisticated than I am describing here, are going to have a hard time deciding whether they are conscious in the human sense. Since humans seem essentially unable to describe consciousness in any reasonable operational terms, there doesn't seem any acceptable way for an AI to decide whether the word applies to itself. And given this failure, it calls into question the ease with which humans assert that they are conscious. How do we really know that we are conscious? For example, how do we know that what we call consciousness is what everyone else calls consciousness? I am worried that many people believe they are conscious simply because as children, they were told they were conscious. They were told that consciousness is the difference between being awake and being asleep, and assume on that basis that when they are awake they are conscious. Then all those other synonyms are treated the same way. Yet most humans would not admit to any doubt that they are conscious. For such a slippery and seemingly undefinable concept, it seems odd that people are so sure of it. Why, then, can't an AI achieve a similar degree of certainty? Do you think a properly programmed AI would ever say, yes, I am conscious, because I have subjectivity, self-awareness, sentience, sapience, etc., and I know this because it is just inherent in my artificial brain? Presumably we could program the AI to say this, and to believe it (in whatever sense that word applies), but is it something an AI could logically conclude? Hal --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups Everything List group. To post to this group, send email to [EMAIL PROTECTED] To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/everything-list?hl=en -~--~~~~--~~--~--~---
Re: How would a computer know if it were conscious?
Consciousness is a cognitive system capable of reflecting on other cognitive systems, by enabling switching and integration between differing representations of knowledge in different domains. It's a higher-level summary of knowledge in which there is a degree of coarse graining sufficient to lose precise information about the under-lying computations. Current experience is integrated with past knowledge in order to provide higher-level summaries of the meaning of a concept. Any cognitive system capable of reflection in this sense is conscious. In essence, conscious is what *mediates* between different representations of knowledge... as mentioned above... the ability to switch between and integrate different representational systems. There are three general types of consciousness arising from the fact that there are three different classes of cognitive systems which could be potentially reflected upon. The first are systems which perceive physical concepts. When this perception is reflected upon, we experience sensations. The second are systems which perceive teleological concepts... closely related to our motivational systems. When this is reflected upon, we experience emotions (or more accurately feelings). The third type of consciousness is very weak in humans... it's the ability to reflect upon systems which perceive logical/mathematical things. Reflection upon these systems is consciously experienced as an 'ontology-scape' (in a sense, conscious awareness of the theory of everything). But as mentioned, this last type of consciousness is very weak in humans, since our ability to reflect upon our own cognitive systems is quite small and not done by the brain directly (when engaged in logical reasoning, we humans are not generally reflecting on our thoughts directly, but via indirect means such as verbal or visual representations of these thoughts). The third type of conscious mentioned above is synonymous with 'reflective intelligence'. That is, any system successfully engaged in reflective decision theory would automatically be conscious. Incidentally, such a system would also be 'friendly' (ethical) automatically. The ability to reason effectively about ones own cognitive processes would certainly enable the ability to elaborate precise definitions of consciousness and determine that the system was indeed conforming to the aforementioned definitions. Much of the confusion surrounding these issues stems from the fact there's not one definition of 'general intelligence', but THREE. There's the ability to detect patterns (which does not require sentience), there's the ability to engage in symbolic reasoning (which also does not require sentience), finally there's the ability to engage in reflective reasoning (reasoning about reasoning). And it's this third definition intelligence which DOES by necessity entail consciousness. It also, by necessity entails ethical behaviour. Now there those who point to powerful systems such as 'Corporations' and 'Evolution' to try to argue that you can have intelligence without consciousness. But these arguments are not convincing. It's true that, for instance 'Evolution' is an intelligence system in ONE sense, but it's certainly NOT a *reflective intelligence*. Nor, is , for instance a 'Corporation'. A 'corporation' has SOME of the sub-systems of person-hood, but not all of them. The existence of non-sentient cognitive systems which display *some* of the features of intelligence is a LONG way from establishing that you can have *reflective intelligence* without consciousness. As has been pointed out, RPOPS such as for instance 'Evolution' are NOT capable of reflective intelligence and therefore cannot be taken as a disproof of the claim that reflective intelligence must by necessity automatically be friendly and sentient. The reason I elaborate this carefully is because you will, on certain mailing lists, run into a number of artificial intelligence crack-pots claiming that (1) You can have reflective intelligence without consciousness and (2) AI's with reflective intelligence don't have to be ethical and might destroy the world. Ignore these crack-pot claims when you see them. --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups Everything List group. To post to this group, send email to [EMAIL PROTECTED] To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/everything-list?hl=en -~--~~~~--~~--~--~---
Re: How would a computer know if it were conscious?
On 03/06/07, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote: The third type of conscious mentioned above is synonymous with 'reflective intelligence'. That is, any system successfully engaged in reflective decision theory would automatically be conscious. Incidentally, such a system would also be 'friendly' (ethical) automatically. The ability to reason effectively about ones own cognitive processes would certainly enable the ability to elaborate precise definitions of consciousness and determine that the system was indeed conforming to the aforementioned definitions. How do you derive (a) ethics and (b) human-friendly ethics from reflective intelligence? I don't see why an AI should decide to destroy the world, save the world, or do anything at all to the world, unless it started off with axioms and goals which pushed it in a particular direction. -- Stathis Papaioannou --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups Everything List group. To post to this group, send email to [EMAIL PROTECTED] To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/everything-list?hl=en -~--~~~~--~~--~--~---
Re: How would a computer know if it were conscious?
On Jun 3, 9:20 pm, Stathis Papaioannou [EMAIL PROTECTED] wrote:
On 03/06/07, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote:
The third type of conscious mentioned above is synonymous with
'reflective intelligence'. That is, any system successfully engaged
in reflective decision theory would automatically be conscious.
Incidentally, such a system would also be 'friendly' (ethical)
automatically. The ability to reason effectively about ones own
cognitive processes would certainly enable the ability to elaborate
precise definitions of consciousness and determine that the system was
indeed conforming to the aforementioned definitions.
How do you derive (a) ethics and (b) human-friendly ethics from reflective
intelligence? I don't see why an AI should decide to destroy the world,
save the world, or do anything at all to the world, unless it started off
with axioms and goals which pushed it in a particular direction.
--
Stathis Papaioannou
When reflective intelligence is applied to cognitive systems which
reason about teleological concepts (which include values, motivations
etc) the result is conscious 'feelings'. Reflective intelligence,
recall, is the ability to correctly reason about cognitive systems.
When applied to cognitive systems reasoning about teleological
concepts this means the ability to correctly determine the
motivational 'states' of self and others - as mentioned - doing this
rapidly and accuracy generates 'feelings'. Since, as has been known
since Hume, feelings are what ground ethics, the generation of
feelings which represent accurate tokens about motivational
automatically leads to ethical behaviour.
Bad behaviour in humans is due to a deficit in reflective
intelligence. It is known for instance, that psychopaths have great
difficulty perceiving fear and sadness and negative motivational
states in general. Correct representation of motivational states is
correlated with ethical behaviour. Thus it appears that reflective
intelligence is automatically correlated with ethical behaviour. Bear
in mind, as I mentioned that: (1) There are in fact three kinds of
general intelligence, and only one of them ('reflective intelligence')
is correlated with ethics.The other two are not. A deficit in
reflective intelligence does not affect the other two types of general
intelligence (which is why for instance psychopaths could still score
highly in IQ tests). And (2) Reflective intelligence in human beings
is quite weak. This is the reason why intelligence does not appear to
be much correlated with ethics in humans. But this fact in no way
refutes the idea that a system with full and strong reflective
intelligence would automatically be ethical.
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups
Everything List group.
To post to this group, send email to [EMAIL PROTECTED]
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at
http://groups.google.com/group/everything-list?hl=en
-~--~~~~--~~--~--~---
Re: How would a computer know if it were conscious?
On 03/06/07, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote:
How do you derive (a) ethics and (b) human-friendly ethics from reflective
intelligence? I don't see why an AI should decide to destroy the world,
save the world, or do anything at all to the world, unless it started
off
with axioms and goals which pushed it in a particular direction.
When reflective intelligence is applied to cognitive systems which
reason about teleological concepts (which include values, motivations
etc) the result is conscious 'feelings'. Reflective intelligence,
recall, is the ability to correctly reason about cognitive systems.
When applied to cognitive systems reasoning about teleological
concepts this means the ability to correctly determine the
motivational 'states' of self and others - as mentioned - doing this
rapidly and accuracy generates 'feelings'. Since, as has been known
since Hume, feelings are what ground ethics, the generation of
feelings which represent accurate tokens about motivational
automatically leads to ethical behaviour.
Determining the motivational states of others does not necessarily involve
feelings or empathy. It has been historically very easy to assume that other
species or certain members of our own species either lack feelings or, if
they have them, it doesn't matter. Moreover, this hasn't prevented people
from determining the motivations of inferior beings in order to exploit
them. So although having feelings may be necessary for ethical behaviour, it
is not sufficient.
Bad behaviour in humans is due to a deficit in reflective
intelligence. It is known for instance, that psychopaths have great
difficulty perceiving fear and sadness and negative motivational
states in general. Correct representation of motivational states is
correlated with ethical behaviour.
Psychopaths are often very good at understanding other peoples' feelings, as
evidenced by their ability to manipulate them. The main problem is that they
don't *care* about other people; they seem to lack the ability to be moved
by other peoples' emotions and lack the ability to experience emotions such
as guilt. But this isn't part of a general inability to feel emotion, as
they often present as enraged, entitled, depressed, suicidal, etc., and
these emotions are certainly enough to motivate them. Psychopaths have a
slightly different set of emotions, regulated in a different way compared to
the rest of us, but are otherwise cognitively intact.
Thus it appears that reflective
intelligence is automatically correlated with ethical behaviour. Bear
in mind, as I mentioned that: (1) There are in fact three kinds of
general intelligence, and only one of them ('reflective intelligence')
is correlated with ethics.The other two are not. A deficit in
reflective intelligence does not affect the other two types of general
intelligence (which is why for instance psychopaths could still score
highly in IQ tests). And (2) Reflective intelligence in human beings
is quite weak. This is the reason why intelligence does not appear to
be much correlated with ethics in humans. But this fact in no way
refutes the idea that a system with full and strong reflective
intelligence would automatically be ethical.
Perhaps I haven't quite understood your definition of reflective
intelligence. It seems to me quite possible to correctly reason about
cognitive systems, at least well enough to predict their behaviour to a
useful degree, and yet not care at all about what happens to them.
Furthermore, it seems possible to me to do this without even suspecting that
the cognitive system is conscious, or at least without being sure that it is
conscious.
--
Stathis Papaioannou
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups
Everything List group.
To post to this group, send email to [EMAIL PROTECTED]
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at
http://groups.google.com/group/everything-list?hl=en
-~--~~~~--~~--~--~---
Re: How would a computer know if it were conscious?
What do others on this list think about Mark Tegmark's definition of
consciousness:
I believe that consciousness is, essentially, the way information
feels when being processed. Since matter can be arranged to process
information in numerous ways of vastly varying complexity, this
implies a rich variety of levels and types of consciousness.
Source: http://www.edge.org/q2007/q07_7.html
Jason
On Jun 3, 6:11 am, Stathis Papaioannou [EMAIL PROTECTED] wrote:
On 03/06/07, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote:
How do you derive (a) ethics and (b) human-friendly ethics from reflective
intelligence? I don't see why an AI should decide to destroy the world,
save the world, or do anything at all to the world, unless it started
off
with axioms and goals which pushed it in a particular direction.
When reflective intelligence is applied to cognitive systems which
reason about teleological concepts (which include values, motivations
etc) the result is conscious 'feelings'. Reflective intelligence,
recall, is the ability to correctly reason about cognitive systems.
When applied to cognitive systems reasoning about teleological
concepts this means the ability to correctly determine the
motivational 'states' of self and others - as mentioned - doing this
rapidly and accuracy generates 'feelings'. Since, as has been known
since Hume, feelings are what ground ethics, the generation of
feelings which represent accurate tokens about motivational
automatically leads to ethical behaviour.
Determining the motivational states of others does not necessarily involve
feelings or empathy. It has been historically very easy to assume that other
species or certain members of our own species either lack feelings or, if
they have them, it doesn't matter. Moreover, this hasn't prevented people
from determining the motivations of inferior beings in order to exploit
them. So although having feelings may be necessary for ethical behaviour, it
is not sufficient.
Bad behaviour in humans is due to a deficit in reflective
intelligence. It is known for instance, that psychopaths have great
difficulty perceiving fear and sadness and negative motivational
states in general. Correct representation of motivational states is
correlated with ethical behaviour.
Psychopaths are often very good at understanding other peoples' feelings, as
evidenced by their ability to manipulate them. The main problem is that they
don't *care* about other people; they seem to lack the ability to be moved
by other peoples' emotions and lack the ability to experience emotions such
as guilt. But this isn't part of a general inability to feel emotion, as
they often present as enraged, entitled, depressed, suicidal, etc., and
these emotions are certainly enough to motivate them. Psychopaths have a
slightly different set of emotions, regulated in a different way compared to
the rest of us, but are otherwise cognitively intact.
Thus it appears that reflective
intelligence is automatically correlated with ethical behaviour. Bear
in mind, as I mentioned that: (1) There are in fact three kinds of
general intelligence, and only one of them ('reflective intelligence')
is correlated with ethics.The other two are not. A deficit in
reflective intelligence does not affect the other two types of general
intelligence (which is why for instance psychopaths could still score
highly in IQ tests). And (2) Reflective intelligence in human beings
is quite weak. This is the reason why intelligence does not appear to
be much correlated with ethics in humans. But this fact in no way
refutes the idea that a system with full and strong reflective
intelligence would automatically be ethical.
Perhaps I haven't quite understood your definition of reflective
intelligence. It seems to me quite possible to correctly reason about
cognitive systems, at least well enough to predict their behaviour to a
useful degree, and yet not care at all about what happens to them.
Furthermore, it seems possible to me to do this without even suspecting that
the cognitive system is conscious, or at least without being sure that it is
conscious.
--
Stathis Papaioannou
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups
Everything List group.
To post to this group, send email to [EMAIL PROTECTED]
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at
http://groups.google.com/group/everything-list?hl=en
-~--~~~~--~~--~--~---
Re: How would a computer know if it were conscious?
Part of what I wanted to get at in my thought experiment is the bafflement and confusion an AI should feel when exposed to human ideas about consciousness. Various people here have proffered their own ideas, and we might assume that the AI would read these suggestions, along with many other ideas that contradict the ones offered here. It seems hard to escape the conclusion that the only logical response is for the AI to figuratively throw up its hands and say that it is impossible to know if it is conscious, because even humans cannot agree on what consciousness is. In particular I don't think an AI could be expected to claim that it knows that it is conscious, that consciousness is a deep and intrinsic part of itself, that whatever else it might be mistaken about it could not be mistaken about being conscious. I don't see any logical way it could reach this conclusion by studying the corpus of writings on the topic. If anyone disagrees, I'd like to hear how it could happen. And the corollary to this is that perhaps humans also cannot legitimately make such claims, since logically their position is not so different from that of the AI. In that case the seemingly axiomatic question of whether we are conscious may after all be something that we could be mistaken about. Hal --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups Everything List group. To post to this group, send email to [EMAIL PROTECTED] To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/everything-list?hl=en -~--~~~~--~~--~--~---
Re: How would a computer know if it were conscious?
Why would we have a word that intuitively everybody can grasp for himself without it being linked to a real phenomena ? Not only we have one word, but we have plenty of words which try to grasp the idea. Denying consciousness phenomena like this is playing a vocabulary game... not denying the subject of the word. Quentin On Sunday 03 June 2007 21:52:17 Hal Finney wrote: Part of what I wanted to get at in my thought experiment is the bafflement and confusion an AI should feel when exposed to human ideas about consciousness. Various people here have proffered their own ideas, and we might assume that the AI would read these suggestions, along with many other ideas that contradict the ones offered here. It seems hard to escape the conclusion that the only logical response is for the AI to figuratively throw up its hands and say that it is impossible to know if it is conscious, because even humans cannot agree on what consciousness is. In particular I don't think an AI could be expected to claim that it knows that it is conscious, that consciousness is a deep and intrinsic part of itself, that whatever else it might be mistaken about it could not be mistaken about being conscious. I don't see any logical way it could reach this conclusion by studying the corpus of writings on the topic. If anyone disagrees, I'd like to hear how it could happen. And the corollary to this is that perhaps humans also cannot legitimately make such claims, since logically their position is not so different from that of the AI. In that case the seemingly axiomatic question of whether we are conscious may after all be something that we could be mistaken about. Hal --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups Everything List group. To post to this group, send email to [EMAIL PROTECTED] To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/everything-list?hl=en -~--~~~~--~~--~--~---
Re: How would a computer know if it were conscious?
Hal Finney wrote: Part of what I wanted to get at in my thought experiment is the bafflement and confusion an AI should feel when exposed to human ideas about consciousness. Various people here have proffered their own ideas, and we might assume that the AI would read these suggestions, along with many other ideas that contradict the ones offered here. It seems hard to escape the conclusion that the only logical response is for the AI to figuratively throw up its hands and say that it is impossible to know if it is conscious, because even humans cannot agree on what consciousness is. In particular I don't think an AI could be expected to claim that it knows that it is conscious, that consciousness is a deep and intrinsic part of itself, that whatever else it might be mistaken about it could not be mistaken about being conscious. I don't see any logical way it could reach this conclusion by studying the corpus of writings on the topic. If anyone disagrees, I'd like to hear how it could happen. And the corollary to this is that perhaps humans also cannot legitimately make such claims, since logically their position is not so different from that of the AI. In that case the seemingly axiomatic question of whether we are conscious may after all be something that we could be mistaken about. Hal --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups Everything List group. To post to this group, send email to [EMAIL PROTECTED] To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/everything-list?hl=en -~--~~~~--~~--~--~---
Re: How would a computer know if it were conscious?
Sorry about the previous post... I did it from the the Google listsomething weird happened. --- Hi folks, Re: How would a computer know if it were conscious? Easy. The computer would be able to go head to head with a human in a competition. The competition? Do science on exquisite novelty that neither party had encountered. (More interesting: Make their life depend on getting it right. The survivors are conscious). Only conscious entities can do open ended science on the exquisitely novel. You cannot teach something how to deal with the exquisitely novel because you haven't any experience of it to teach. It means that the entity must be configurted as a machine that learns how to learn something. This is one meta-level removed from your usual AI situation. It's what humans do. During neogenesis and development, humans 'learn how to learn how to learn. If the computer/scientist can match the human/scientist...it's as conscious as a human. It must be. cheers colin hales --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups Everything List group. To post to this group, send email to [EMAIL PROTECTED] To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/everything-list?hl=en -~--~~~~--~~--~--~---
Re: How would a computer know if it were conscious?
I don't see that you've made your point. If you achieve this, you have created an artificial creative process, a sort of holy grail of AI/ALife. However, it seems far from obvious that consciousness should be necessary. Biological evolution is widely considered to be creative (even exponentially so), but few would argue that the biosphere is conscious (and has been for ca 4E10 years). Cheers On Mon, Jun 04, 2007 at 10:48:07AM +1000, Colin Hales wrote: Sorry about the previous post... I did it from the the Google listsomething weird happened. --- Hi folks, Re: How would a computer know if it were conscious? Easy. The computer would be able to go head to head with a human in a competition. The competition? Do science on exquisite novelty that neither party had encountered. (More interesting: Make their life depend on getting it right. The survivors are conscious). Only conscious entities can do open ended science on the exquisitely novel. You cannot teach something how to deal with the exquisitely novel because you haven't any experience of it to teach. It means that the entity must be configurted as a machine that learns how to learn something. This is one meta-level removed from your usual AI situation. It's what humans do. During neogenesis and development, humans 'learn how to learn how to learn. If the computer/scientist can match the human/scientist...it's as conscious as a human. It must be. cheers colin hales -- A/Prof Russell Standish Phone 0425 253119 (mobile) Mathematics UNSW SYDNEY 2052 [EMAIL PROTECTED] Australiahttp://www.hpcoders.com.au --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups Everything List group. To post to this group, send email to [EMAIL PROTECTED] To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/everything-list?hl=en -~--~~~~--~~--~--~---
Re: How would a computer know if it were conscious?
If it feels bafflement and confusion, then surely it is conscious :) An AI that takes information from books might experience similar qualia we can experience. The AI will be programmed to do certain tasks and it must thus have a notion of what it is doing is ok., not ok, or completely wrong. If things are going wrong and it has to revert what it has just done, it may feel some sort of pain. Just like what happens to us if we pick up something that is very hot. So, I think that there will be a mismatch between the qualia the AI experiences and what it reads about that we experience. The AI won't read the information like we read it. I think it will directly experience it as some qualia, just like we experience information coming in via our senses into our brain. The meaning we associate with the text would not be accessible to the AI, because ultimately that is linked to the qualia we experience. Perhaps what the AI experiences when it is processing information is similar to an animal that is moving in some landscape. Maybe when it reads something then that manifests itself like some object it sees. If it processes information then that could be like picking up that object putting it next to a similar looking object. But if that object represents a text about consciousness then there is no way for the AI to know that. Saibal - Original Message - From: Hal Finney [EMAIL PROTECTED] To: [EMAIL PROTECTED] Sent: Sunday, June 03, 2007 09:52 PM Subject: Re: How would a computer know if it were conscious? Part of what I wanted to get at in my thought experiment is the bafflement and confusion an AI should feel when exposed to human ideas about consciousness. Various people here have proffered their own ideas, and we might assume that the AI would read these suggestions, along with many other ideas that contradict the ones offered here. It seems hard to escape the conclusion that the only logical response is for the AI to figuratively throw up its hands and say that it is impossible to know if it is conscious, because even humans cannot agree on what consciousness is. In particular I don't think an AI could be expected to claim that it knows that it is conscious, that consciousness is a deep and intrinsic part of itself, that whatever else it might be mistaken about it could not be mistaken about being conscious. I don't see any logical way it could reach this conclusion by studying the corpus of writings on the topic. If anyone disagrees, I'd like to hear how it could happen. And the corollary to this is that perhaps humans also cannot legitimately make such claims, since logically their position is not so different from that of the AI. In that case the seemingly axiomatic question of whether we are conscious may after all be something that we could be mistaken about. Hal --~--~-~--~~~---~--~~ You received this message because you are subscribed to the Google Groups Everything List group. To post to this group, send email to [EMAIL PROTECTED] To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/everything-list?hl=en -~--~~~~--~~--~--~---
