Re: [agi] Ethics of computer-based cognitive experimentation
Trent Waddington wrote: As I believe the is that conciousness? debate could go on forever, I think I should make an effort here to save this thread. Setting aside the objections of vegetarians and animal lovers, many hard nosed scientists decided long ago that jamming things into the brains of monkeys and the like is justifiable treatment of creatures suspected by many to have similar experiences to humans. If you're in agreement with these practices then I think you should be in agreement with any and all experimentation on simulated networks of complexity up to and including these organisms. Yes, my intent on starting this thread was not to define consciousness, but rather to ask how do we make ethical choices with regard to AGI before we are able to define it? I agree with your points above. However, I am not entirely sanguine about animal experiments. I accept that they're sometimes OK, or at least the lesser of two evils, but I would prefer to avoid even that level of compromise when experimenting on AGIs. And, given that we have the ability to design the AGI experimental subject -- as opposed to being stuck with a pre-designed animal -- it /should/ be possible. --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
On Fri, Nov 14, 2008 at 2:07 AM, John G. Rose [EMAIL PROTECTED] wrote: there are many computer systems now, domain specific intelligent ones where their life is more important than mine. Some would say that the battle is already lost. For now, it's not really your life (or interest) vs the system's life (or interest). It's rather your life (or interest) vs lives (or interests) of people the system protects/supports. Our machines still work for humans. At least it still seems to be the case ;-)). If we are stupid enough to develop very powerful machines without equally powerful safety controls then we (just like many other species) are due for extinction for adaptability limitations. Regards, Jiri Jelinek --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
RE: [agi] Ethics of computer-based cognitive experimentation
From: Jiri Jelinek [mailto:[EMAIL PROTECTED] On Fri, Nov 14, 2008 at 2:07 AM, John G. Rose [EMAIL PROTECTED] wrote: there are many computer systems now, domain specific intelligent ones where their life is more important than mine. Some would say that the battle is already lost. For now, it's not really your life (or interest) vs the system's life (or interest). It's rather your life (or interest) vs lives (or interests) of people the system protects/supports. Our machines still work for humans. At least it still seems to be the case ;-)). If we are stupid enough to develop very powerful machines without equally powerful safety controls then we (just like many other species) are due for extinction for adaptability limitations. It is where the interests of others is more valuable than an individual's life. Ancient Rome had the entertainment interests of the masses at a higher value than those being devoured by lions in the arena. I would say that computers and machines interests today in many cases now are of similar relational circumstances in some cases. Our herd mentality makes it easy for rights to be taken away and at the same time it is accepted and defended as necessary and an improvement. Example - anonymity and privacy = gone. Sounds paranoiacal but there are many that agree on this. It is an icky subject, easy to ignore, and perhaps something that hinders technological progression. John --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
Why consciousness is hard to define (was Re: [agi] Ethics of computer-based cognitive experimentation)
--- On Fri, 11/14/08, Colin Hales [EMAIL PROTECTED] wrote: Try running yourself with empirical results instead of metabelief (belief about belief). You'll get someplace .i.e. you'll resolve the inconsistencies. When inconsistencies are testably absent, no matter how weird the answer, it will deliver maximally informed choices. Not facts. Facts will only ever appear differently after choices are made. This too is a fact...which I have chosen to make choices about. :-) If you fail to resolve your inconsistency then you are guaranteeing that your choices are minimally informed. Fine. By your definition of consciousness, I must be conscious because I can see and because I can apply the scientific method, which you didn't precisely define, but I assume that means I can do experiments and learn from them. But by your definition, a simple modification to autobliss ( http://www.mattmahoney.net/autobliss.txt ) would make it conscious. It already applies the scientific method. It outputs 3 bits (2 randomly picked inputs to an unknown logic gate and a proposed output) and learns the logic function. The missing component is vision. But suppose I replace the logic function (a 4 bit value specified by the teacher) with a black box with 3 switches and a light bulb to indicate whether the proposed output (one of the switches) is right or wrong. You also didn't precisely define what constitutes vision, so I assume a 1 pixel system qualifies. Of course I don't expect anyone to precisely define consciousness (as a property of Turing machines). There is no algorithmically simple definition that agrees with intuition, i.e. that living humans and nothing else are conscious. This goes beyond Rice's theorem, which would make any nontrivial definition not computable. Even allowing non computable definitions (the output can be yes, no, or maybe), you still have the problem that any specification with algorithmic complexity K can be expressed as a program with complexity K. Given any simple specification (meaning K is small) I can write a simple program that satisfies it (my program has complexity at most K). However, for humans, K is about 10^9 bits. That means any specification smaller than a 1 GB file or 1000 books would allow a counter intuitive example of a simple program that meets your test for consciousness. Try it if you don't believe me. Give me a simple definition of consciousness without pointing to a human (like the Turing test does). I am looking for a program is_conscious(x) shorter than 10^9 bits that inputs a Turing machine x and outputs yes, no, or maybe. -- Matt Mahoney, [EMAIL PROTECTED] --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
--- On Wed, 11/12/08, Harry Chesley [EMAIL PROTECTED] wrote: Matt Mahoney wrote: If you don't define consciousness in terms of an objective test, then you can say anything you want about it. We don't entirely disagree about that. An objective test is absolutely crucial. I believe where we disagree is that I expect there to be such a test one day, while you claim there can never be. It depends on the definition. The problem with the current definition (what most people think it means) is that it leads to logical inconsistencies. I believe I have a consciousness, a little person inside my head that experiences things and makes decisions. I also believe that my belief is false, that my brain would do exactly the same thing without this little person. I know these two views are inconsistent. I just accept that they are and leave it at that. -- Matt Mahoney, [EMAIL PROTECTED] --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
Dear Matt, Try running yourself with empirical results instead of metabelief (belief about belief). You'll get someplace .i.e. you'll resolve the inconsistencies. When inconsistencies are *testably *absent, no matter how weird the answer, it will deliver maximally informed choices. Not facts. Facts will only ever appear differently after choices are made. This too is a fact...which I have chosen to make choices about. :-) If you fail to resolve your inconsistency then you are guaranteeing that your choices are minimally informed. Tricky business, science: an intrinsically dynamic process in which choice is the driver (epistemic state transition) and the facts (the epistemic state) are forever transitory , never certain. You can only make so-called facts certain by failing to choose. Then they lodge in your brain (and nowhere else) like dogma-crud between your teeth, and the rot sets in. The plus side - you get to be 100% right. Personally I'd rather get real AGI built and be testably wrong a million times along the way. cheers, colin hales Matt Mahoney wrote: --- On Wed, 11/12/08, Harry Chesley [EMAIL PROTECTED] wrote: Matt Mahoney wrote: If you don't define consciousness in terms of an objective test, then you can say anything you want about it. We don't entirely disagree about that. An objective test is absolutely crucial. I believe where we disagree is that I expect there to be such a test one day, while you claim there can never be. It depends on the definition. The problem with the current definition (what most people think it means) is that it leads to logical inconsistencies. I believe I have a consciousness, a little person inside my head that experiences things and makes decisions. I also believe that my belief is false, that my brain would do exactly the same thing without this little person. I know these two views are inconsistent. I just accept that they are and leave it at that. -- Matt Mahoney, [EMAIL PROTECTED] --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?; Powered by Listbox: http://www.listbox.com --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
As I believe the is that conciousness? debate could go on forever, I think I should make an effort here to save this thread. Setting aside the objections of vegetarians and animal lovers, many hard nosed scientists decided long ago that jamming things into the brains of monkeys and the like is justifiable treatment of creatures suspected by many to have similar experiences to humans. If you're in agreement with these practices then I think you should be in agreement with any and all experimentation on simulated networks of complexity up to and including these organisms. I, personally, say that these experiments are just fine because lab animals are property and I have more respect for property rights than I do for save the animals causes. Even if you doubt that lab animals are property, I don't expect that your doubts will extend to computer hardware, whether or not there is a sophisticated simulation running on it or not. Trent --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
RE: [agi] Ethics of computer-based cognitive experimentation
From: Jiri Jelinek [mailto:[EMAIL PROTECTED] On Wed, Nov 12, 2008 at 2:41 AM, John G. Rose [EMAIL PROTECTED] wrote: is it really necessary for an AGI to be conscious? Depends on how you define it. If you think it's about feelings/qualia then - no - you don't need that [potentially dangerous] crap + we don't know how to implement it anyway. If you view it as high-level built-in response mechanism (which is supported by feelings in our brain but can/should be done differently in AGI) then yes - you practically (but not necessarily theoretically) need something like that for performance. If you are concerned about self-awareness/consciousness then note that AGI can demonstrate general problem solving without knowing anything about itself (and about many other particular concepts). The AGI just should be able to learn new concepts (including self), though I think some built-in support makes sense in this particular case. BTW for the purpose of my AGI RD I defined self-awareness as a use of an internal representation (IR) of self, where the IR is linked to real features of the system. Nothing terribly complicated or mysterious about that. Yes, I agree that problem solving can be performed without self-awareness and I believe that actions involving rich intelligence need not require consciousness. But yes it all depends on how you define consciousness. It can be argued that a rock is conscious. Doesn't that complicate things? it does Shouldn't the machines/computers be slaves to man? They should and it shouldn't be viewed negatively. It's nothing more than a smart tool. Changing that would be a big mistake IMO. Yup when you need to scuttle the spaceship and HAL is having issues with that uhm it would be better for HAL to understand that he is expendable. Though there are AGI applications that would involve humans building close interpersonal relationships for various reasons. I mean having that AGI psychotherapist could be useful :) And advanced post-Singularity AGI applications, yes, I suppose machine consciousness and consciousness uploading and mixing, ya, in the meantime though for pre-Singularity design and study I don't see machine consciousness as required, human equiv that is. Though I do have a fuzzy view of how I would design a consciousness. Or will they be equal/superior. Rocks are superior to us in being hard. Cars are superior to us when it comes to running fast. AGIs will be superior to us when it comes to problem solving. So what? Equal/superior in whatever - who cares as long as we can progress safely enjoy life - which is what our tools (including AGI) are being designed to help us with. Superior meaning - if it was me or AGI-X due to limited resources does AGI-X get to live and I am expendable. Unfortunately there are many computer systems now, domain specific intelligent ones where their life is more important than mine. Some would say that the battle is already lost. John --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
RE: [agi] Ethics of computer-based cognitive experimentation
From: Richard Loosemore [mailto:[EMAIL PROTECTED] I thought what he said was a good description more or less. Out of 600 millions years there may be only a fraction of that which is an improvement but it's still there. How do you know, beyond a reasonable doubt, that any other being is conscious? The problem is, you have to nail down exactly what you *mean* by the word conscious before you start asking questions or making statements. Once you start reading about and thinking about all the attempts that have been made to get specific about it, some interesting new answers to simple questions like this begin to emerge. What I am fighting here is a tendency for some people to use wave-of-the-hand definitions that only capture a fraction of a percent of the real meaning of the term. And sometimes not even that. I see consciousness as a handle to a system. Consciousness is and is not a unit. Being a system it has components. And the word consciousness may be semi-inclusive or over-inclusive. As well consciousness can be descripted as an ether type thing also but consciousness as a system is more applicable here I think. I would be interested in how one goes about proving that another being is conscious. I can imagine definitions of consciousness that would prove that. Somehow though the mystery is worthy of perpetuation. One of the main conclusions of the paper I am writing now is that you will (almost certainly) have no choice in the matter, because a sufficiently powerful type of AGI will be conscious whether you like it or not. Uhm what is sufficiently mean here? Consciousness may require some intelligence but I think that intelligence need only possess absolute minimalistic consciousness. Definitions, definitions. Is there someone who has come up with a consciousness system described quantitatively instead of just fussy word descriptions? The question of slavery is completely orthogonal. Yes and no. It's related. I just want things to be taken care of and no issues. Consciousness brings issues. Intelligence and consciousness are separate. Back to my first paragraph above: until you have thought carefully about what you mean by consciousness, and have figured out where it comes from, you can't really make a definitive statement like that, surely? Have deeply thought about it. They are not mutually exclusive nor mostly the same. With both I assume calculations involving resource processing and space time dynamics. Consciousness needs to be broken up into different kinds of consciousness with interrelatedness between. Intelligence has less complexity than consciousness. It is a semi-system. Consciousness can be evoked using intelligence. Intelligence can be spurred with consciousness. They both interoperate but intelligence can be distilled out of an existing conscio-intelligence. And they can facilitate each other yet hinder each other. We'd really have to get into the math to get commitant on it. And besides, the wanting to have things taken care of bit is a separate issue. That is not a problem, either way. Heh. John --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
--- On Tue, 11/11/08, Richard Loosemore [EMAIL PROTECTED] wrote: Matt Mahoney wrote: --- On Tue, 11/11/08, Richard Loosemore [EMAIL PROTECTED] wrote: Your 'belief' explanation is a cop-out because it does not address any of the issues that need to be addressed for something to count as a definition or an explanation of the facts that need to be explained. As I explained, animals that have no concept of death have nevertheless evolved to fear most of the things that can kill them. Humans have learned to associate these things with death, and invented the concept of consciousness as the large set of features which distinguishes living humans from dead humans. Thus, humans fear the loss or destruction of consciousness, which is equivalent to death. Consciousness, free will, qualia, and good and bad are universal human beliefs. We should not confuse them with truth by asking the wrong questions. Thus, Turing sidestepped the question of can machines think? by asking instead can machines appear to think? Since we can't (by definition) distinguish doing something from appearing to do something, it makes no sense for us to make this distinction. The above two paragraphs STILL do not address any of the issues that need to be addressed for something to count as a definition, or an explanation of the facts that need to be explained. And you STILL have not defined what consciousness is. -- Matt Mahoney, [EMAIL PROTECTED] --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
--- On Tue, 11/11/08, Colin Hales [EMAIL PROTECTED] wrote: I'm inclined to agree - this will be an issue in the future... if you have a robot helper and someone comes by and beats it to death in front of your kids, who have some kind of attachment to it...a relationship... then crime (i) may be said to be the psychological damage to the children. Crime (ii) is then the murder and whatever one knows of suffering inflicted on the robot helper. Ethicists are gonna have all manner of novelty to play with. Crime (i) is like when a child's favorite puppy is killed in from of them. Yet children that grow up on farms or around hunting regularly see animals killed and make the necessary emotional adjustment of not attributing consciousness to the victim. An important component of this adjustment is to not give the victim a name. In some African cultures with a high infant mortality rate, it is customary not to name babies until their first birthday. One may wonder if people would develop emotional attachments to machines, like that of the fictional Will Robinson to the robot on Lost in Space, or the actual but weaker attachment of subjects to ELIZA. It is certainly possible. But history suggests we can make the reverse detachment no matter how closely the victims resemble the aggressors. Example include slavery, the Holocaust, Pol Pot, and genocides in Rwanda, Sudan, and eastern Congo. The basic traits responsible for this behavior are in all of us. http://en.wikipedia.org/wiki/Stanford_prison_experiment Unlike crime (i) which can be experimentally measured, crime (ii) is a matter of opinion. -- Matt Mahoney, [EMAIL PROTECTED] --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
This thread has gone back and forth several times concerning the reality of consciousness. So at the risk of extending it further unnecessarily, let me give my view, which seems self-evident to me, but I'm sure isn't to others (meaning they may reasonably disagree with me, not that they're idiots (though I'm open to that possibility too)). 1) I'm talking about the hard question of consciousness. 2) It is real, as it clearly influences our thoughts. On the other hand, though it feels subjectively like it is qualitatively different from other aspects of the world, it probably isn't (but I'm open to being wrong here). 3) We cannot currently define or measure it, but some day we will. 4) Until that day comes, it's really hard to have a non-trivial discussion of it, and too easy to fly off into wild theories concerning it. An analogy: How do you know that humans have blood flowing through their veins? Looking at them, you can't tell. Dissecting them after death, you can't tell -- they have blood, but it's not moving. Cutting them while alive produces spurts of blood, but that could be just because the body is generally pressurized, not because there's any on-going flow through the veins. It requires observing the internals of the body while alive to determine that blood actually flows all the time. And it also helps a lot to have a model of the circulatory system that includes the heart as a pump, etc. With consciousness, we're at the pre-scientific stage, because we know so little about cognition that we're not yet able to open it up and observe it as it operates. This will change, hopefully in my lifetime. --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
John G. Rose wrote: From: Richard Loosemore [mailto:[EMAIL PROTECTED] John LaMuth wrote: Reality check *** Consciousness is an emergent spectrum of subjectivity spanning 600 mill. years of evolution involving mega-trillions of competing organisms, probably selecting for obscure quantum effects/efficiencies Our puny engineering/coding efforts could never approach this - not even in a million years. An outwardly pragmatic language simulation, however, is very do-able. John LaMuth It is not. And we can. I thought what he said was a good description more or less. Out of 600 millions years there may be only a fraction of that which is an improvement but it's still there. How do you know, beyond a reasonable doubt, that any other being is conscious? The problem is, you have to nail down exactly what you *mean* by the word conscious before you start asking questions or making statements. Once you start reading about and thinking about all the attempts that have been made to get specific about it, some interesting new answers to simple questions like this begin to emerge. What I am fighting here is a tendency for some people to use wave-of-the-hand definitions that only capture a fraction of a percent of the real meaning of the term. And sometimes not even that. At some point you have to trust that others are conscious, in the same species, you bring them into your recursive loop of consciousness component mix. A primary component of consciousness is a self definition. Conscious experience is unique to the possessor. It is more than a belief that the possessor herself is conscious but others who appear conscious may be just that, appearing to be conscious. Though at some point there is enough feedback between individuals and/or a group to share consciousness experience. Still though, is it really necessary for an AGI to be conscious? Except for delivering warm fuzzies to the creators? Doesn't that complicate things? Shouldn't the machines/computers be slaves to man? Or will they be equal/superior. It's a dog-eat-dog world out there. One of the main conclusions of the paper I am writing now is that you will (almost certainly) have no choice in the matter, because a sufficiently powerful type of AGI will be conscious whether you like it or not. The question of slavery is completely orthogonal. I just want things to be taken care of and no issues. Consciousness brings issues. Intelligence and consciousness are separate. Back to my first paragraph above: until you have thought carefully about what you mean by consciousness, and have figured out where it comes from, you can't really make a definitive statement like that, surely? And besides, the wanting to have things taken care of bit is a separate issue. That is not a problem, either way. Richard Loosemore --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
Matt Mahoney wrote: --- On Tue, 11/11/08, Richard Loosemore [EMAIL PROTECTED] wrote: Matt Mahoney wrote: --- On Tue, 11/11/08, Richard Loosemore [EMAIL PROTECTED] wrote: Your 'belief' explanation is a cop-out because it does not address any of the issues that need to be addressed for something to count as a definition or an explanation of the facts that need to be explained. As I explained, animals that have no concept of death have nevertheless evolved to fear most of the things that can kill them. Humans have learned to associate these things with death, and invented the concept of consciousness as the large set of features which distinguishes living humans from dead humans. Thus, humans fear the loss or destruction of consciousness, which is equivalent to death. Consciousness, free will, qualia, and good and bad are universal human beliefs. We should not confuse them with truth by asking the wrong questions. Thus, Turing sidestepped the question of can machines think? by asking instead can machines appear to think? Since we can't (by definition) distinguish doing something from appearing to do something, it makes no sense for us to make this distinction. The above two paragraphs STILL do not address any of the issues that need to be addressed for something to count as a definition, or an explanation of the facts that need to be explained. And you STILL have not defined what consciousness is. Logically, I don't need to define something to point out that your definition fails to address any of the issues that I can read about in e.g. Chalmers' book on the subject. ;-) Richard Loosemore --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
Jiri Jelinek wrote: On Wed, Nov 12, 2008 at 2:41 AM, John G. Rose [EMAIL PROTECTED] wrote: is it really necessary for an AGI to be conscious? Depends on how you define it. H interesting angle. Everything you say from this point on seems to be predicated on the idea that a person can *choose* to define it any way they want, and then run with their definition. I notice that this is not possible with any other scientific concept - we don't just define an electron as Your Plastic Pal Who's Fun To Be With and then start drawing conclusions. The same is true of consciousness. Richard Loosemore If you think it's about feelings/qualia then - no - you don't need that [potentially dangerous] crap + we don't know how to implement it anyway. If you view it as high-level built-in response mechanism (which is supported by feelings in our brain but can/should be done differently in AGI) then yes - you practically (but not necessarily theoretically) need something like that for performance. If you are concerned about self-awareness/consciousness then note that AGI can demonstrate general problem solving without knowing anything about itself (and about many other particular concepts). The AGI just should be able to learn new concepts (including self), though I think some built-in support makes sense in this particular case. BTW for the purpose of my AGI RD I defined self-awareness as a use of an internal representation (IR) of self, where the IR is linked to real features of the system. Nothing terribly complicated or mysterious about that. Doesn't that complicate things? it does Shouldn't the machines/computers be slaves to man? They should and it shouldn't be viewed negatively. It's nothing more than a smart tool. Changing that would be a big mistake IMO. Or will they be equal/superior. Rocks are superior to us in being hard. Cars are superior to us when it comes to running fast. AGIs will be superior to us when it comes to problem solving. So what? Equal/superior in whatever - who cares as long as we can progress safely enjoy life - which is what our tools (including AGI) are being designed to help us with. Regards, Jiri Jelinek --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?; Powered by Listbox: http://www.listbox.com --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
Richard, Everything you say from this point on seems to be predicated on the idea that a person can *choose* to define it any way they want There are some good-to-stick-with rules for definitions http://en.wikipedia.org/wiki/Definition#Rules_for_definition_by_genus_and_differentia but (even though it's not desirable) in some cases it's IMO ok for researchers to use a bit different definitions. If you can give us the *ultimate* definition of consciousness then I would certainly be interested. I promise I'll not ask for the ultimate cross-domain definition of every single word used in that definition ;-) Regards, Jiri On Wed, Nov 12, 2008 at 12:16 PM, Richard Loosemore [EMAIL PROTECTED] wrote: Jiri Jelinek wrote: On Wed, Nov 12, 2008 at 2:41 AM, John G. Rose [EMAIL PROTECTED] wrote: is it really necessary for an AGI to be conscious? Depends on how you define it. H interesting angle. Everything you say from this point on seems to be predicated on the idea that a person can *choose* to define it any way they want, and then run with their definition. I notice that this is not possible with any other scientific concept - we don't just define an electron as Your Plastic Pal Who's Fun To Be With and then start drawing conclusions. The same is true of consciousness. Richard Loosemore --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
Jiri Jelinek wrote: Richard, Everything you say from this point on seems to be predicated on the idea that a person can *choose* to define it any way they want There are some good-to-stick-with rules for definitions http://en.wikipedia.org/wiki/Definition#Rules_for_definition_by_genus_and_differentia but (even though it's not desirable) in some cases it's IMO ok for researchers to use a bit different definitions. If you can give us the *ultimate* definition of consciousness then I would certainly be interested. I promise I'll not ask for the ultimate cross-domain definition of every single word used in that definition ;-) Hey, no problem, but I'm now embarrassed and in an awkward position, because I am literally trying to do that. I am trying to sort the problem out once and for all. I am finishing it for submission to AGI-09, so it will be done, ready or not, by the end of today. This is something I started as a student essay in 1986, but I have been trying to nail down a testable prediction that can be applied today, rather than in 20 years time. I do have testable predictions, but not ones that can be tested today, alas. As for the question about definitions, sure, it is true that the rules are not cut in stone for how to do it. It's just that consciousness is a rats nest of conflicting definitions Richard Loosemore --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
--- On Wed, 11/12/08, Harry Chesley [EMAIL PROTECTED] wrote: 1) I'm talking about the hard question of consciousness. 2) It is real, as it clearly influences our thoughts. On the other hand, though it feels subjectively like it is qualitatively different from other aspects of the world, it probably isn't (but I'm open to being wrong here). The correct statement is that you believe it is real. Everybody does. Those who didn't, did not pass on their DNA. 3) We cannot currently define or measure it, but some day we will. You can define it any time you want, or use the existing common definition. The real problem is that the existing definitions lead to absurd conclusions, like Chalmer's fading qualia argument. To avoid logical inconsistencies, you either have to accept that machines that pass the Turing test have experience or qualia (because there is no test to detect qualia), or that qualia does not exist. The latter would be the logical conclusion, except that it conflicts with a belief that is hard-coded into all human brains. 4) Until that day comes, it's really hard to have a non-trivial discussion of it, and too easy to fly off into wild theories concerning it. An analogy: How do you know that humans have blood flowing through their veins? Looking at them, you can't tell. Dissecting them after death, you can't tell -- they have blood, but it's not moving. Cutting them while alive produces spurts of blood, but that could be just because the body is generally pressurized, not because there's any on-going flow through the veins. It requires observing the internals of the body while alive to determine that blood actually flows all the time. And it also helps a lot to have a model of the circulatory system that includes the heart as a pump, etc. Blood flow can be directly observed, for example, by x-rays during an angioplasty. But that isn't the point. Even without direct observation, blood flow is supported by a lot of indirect evidence, for example, when you inject a drug into a vein it quickly spreads to other parts of the body. Even theories for which evidence is harder to observe, for example, the existence of fractional electric charges in quarks, are accepted because the theory makes predictions that can be tested. But there are absolutely no testable predictions that can be made from a theory of consciousness. -- Matt Mahoney, [EMAIL PROTECTED] --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
- Original Message - From: John G. Rose [EMAIL PROTECTED] To: agi@v2.listbox.com Sent: Tuesday, November 11, 2008 11:41 PM Subject: RE: [agi] Ethics of computer-based cognitive experimentation I thought what he said was a good description more or less. Out of 600 millions years there may be only a fraction of that which is an improvement but it's still there. ## Yes, the forebrain has consistently expanded over this timeframe. www.forebrain.org In parallel with this physical (neuroanatomical) certainty is the emergent concomittant refinement of consciousness One could say that each reciprocally drives the other in a mutual survival sense w/ consc. perhaps emerging through a systematic evolutionary refinement of quantum effects (info-entropy -- entanglement -- the list is lengthy)... The biological (refined through evolution) remains the key... John LaMuth ### I just want things to be taken care of and no issues. Consciousness brings issues. Intelligence and consciousness are separate. ### Very well put ! John LaMuth --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?; Powered by Listbox: http://www.listbox.com --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
- Original Message - From: Richard Loosemore [EMAIL PROTECTED] To: agi@v2.listbox.com Sent: Wednesday, November 12, 2008 9:05 AM Subject: Re: [agi] Ethics of computer-based cognitive experimentation One of the main conclusions of the paper I am writing now is that you will (almost certainly) have no choice in the matter, because a sufficiently powerful type of AGI will be conscious whether you like it or not. Richard Loosemore ## Consciousness has only been demonstrated in biological systems. Until we can understand how consc. emerges within biology / neuroanatomy contexts then your AGI assertions amount to nothing but faith-based conjectures ... John LaMuth www.ethicalvalues.com --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
On Wed, Nov 12, 2008 at 2:41 AM, John G. Rose [EMAIL PROTECTED] wrote: is it really necessary for an AGI to be conscious? Depends on how you define it. If you think it's about feelings/qualia then - no - you don't need that [potentially dangerous] crap + we don't know how to implement it anyway. If you view it as high-level built-in response mechanism (which is supported by feelings in our brain but can/should be done differently in AGI) then yes - you practically (but not necessarily theoretically) need something like that for performance. If you are concerned about self-awareness/consciousness then note that AGI can demonstrate general problem solving without knowing anything about itself (and about many other particular concepts). The AGI just should be able to learn new concepts (including self), though I think some built-in support makes sense in this particular case. BTW for the purpose of my AGI RD I defined self-awareness as a use of an internal representation (IR) of self, where the IR is linked to real features of the system. Nothing terribly complicated or mysterious about that. Doesn't that complicate things? it does Shouldn't the machines/computers be slaves to man? They should and it shouldn't be viewed negatively. It's nothing more than a smart tool. Changing that would be a big mistake IMO. Or will they be equal/superior. Rocks are superior to us in being hard. Cars are superior to us when it comes to running fast. AGIs will be superior to us when it comes to problem solving. So what? Equal/superior in whatever - who cares as long as we can progress safely enjoy life - which is what our tools (including AGI) are being designed to help us with. Regards, Jiri Jelinek --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
Richard Your proposal sounds very deep ... I look forward to reading it. Please consider adding (to your citations) my own contribution to the field ... LaMuth, J. E. (1977). The Development of the Forebrain as an Elementary Function of the Parameters of Input Specificity and Phylogenetic Age. J. U-grad Rsch: Bio. Sci. U. C. Irvine. (6): 274-294. (as reproduced at ...) http://www.angelfire.com/rnb/fairhaven/brainresearch.html Cordially John LaMuth www.forebrain.org # - Original Message - From: Richard Loosemore [EMAIL PROTECTED] To: agi@v2.listbox.com Sent: Wednesday, November 12, 2008 1:36 PM Subject: Re: [agi] Ethics of computer-based cognitive experimentation John LaMuth wrote: - Original Message - From: Richard Loosemore [EMAIL PROTECTED] To: agi@v2.listbox.com Sent: Wednesday, November 12, 2008 9:05 AM Subject: Re: [agi] Ethics of computer-based cognitive experimentation One of the main conclusions of the paper I am writing now is that you will (almost certainly) have no choice in the matter, because a sufficiently powerful type of AGI will be conscious whether you like it or not. Richard Loosemore ## Consciousness has only been demonstrated in biological systems. Until we can understand how consc. emerges within biology / neuroanatomy contexts then your AGI assertions amount to nothing but faith-based conjectures ... Actually, my proposal is not just about AGI, as you imply, it is about understanding how consciousness emerges within biology/neuroanatomy (more generally, how it emerges within any system that shows intelligence of a cerrtain sort, regardless of substrate). So it applies equally to the biology and the AGI cases. I would never suggest a solution to the problem if it did not cover both. Richard Loosemore --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?; Powered by Listbox: http://www.listbox.com --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
Matt Mahoney wrote: 2) It is real, as it clearly influences our thoughts. On the other hand, though it feels subjectively like it is qualitatively different from other aspects of the world, it probably isn't (but I'm open to being wrong here). The correct statement is that you believe it is real. Everybody does. Those who didn't, did not pass on their DNA. No, the correct statement is the one I made. It is real. We have empirical evidence that it is real since it influences observable actions. Consciousness *may* be a belief. But we have no empirical evidence for or against that statement, so it's too early to make blanket statements like yours. 3) We cannot currently define or measure it, but some day we will. You can define it any time you want, or use the existing common definition. No, you can't define it any way you want. I am talking about a specific phenomenon that has been observed but not understood. And the definitions from others that I've seen may allow us to identify shared experiences of the phenomenon, but don't provide either a good model or empirical tests, so they're less that I, for one, want in order to say we've defined it. Blood flow can be directly observed, for example, by x-rays during an angioplasty. But that isn't the point. Even without direct observation, blood flow is supported by a lot of indirect evidence, for example, when you inject a drug into a vein it quickly spreads to other parts of the body. Even theories for which evidence is harder to observe, for example, the existence of fractional electric charges in quarks, are accepted because the theory makes predictions that can be tested. So far we're in complete agreement. Concluding that blood flows requires observation which requires technology applicable to the phenomenon (x-rays, needles, tests to see if the drug spread, etc.). But there are absolutely no testable predictions that can be made from a theory of consciousness. But here you suddenly jump from saying we have no empirical tests to saying there can be no empirical tests. This makes no sense to me. Even if consciousness is only a belief with no real substance, there are testable predictions that follow from its existence, and perhaps tests to determine that it is limited to being only a belief. --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
Matt Mahoney wrote: snip ... accepted because the theory makes predictions that can be tested. But there are absolutely no testable predictions that can be made from a theory of consciousness. -- Matt Mahoney, [EMAIL PROTECTED] This is simply wrong. It is difficult but you can test for it objectively by demanding that an entity based on your 'theory of consciousness' deliver an authentic scientific act on the a-priori unknown using visual experience for scientific evidence. To the best _indirect_ evidence we have, that act is critically dependent on the existence of a visual phenomenal field within the tested entity. Visual P-consciousness and scientific evidence are literal identities in that circumstance. Degrade visual experience...scientific outcome is disrupted. You can use this to actually discover the physics of qualia as follows: 1) Concoct your theory of consciousness. 2) Build a scientist with it with (amongst other necessities) visual phenomenal consciousness which you believe to be there because of your theory of consciousness. Only autonomous, embodied entities are valid, because it involved actually interacting with an environment the way humans do. 3) Test it for delivery of an authentic act of science on the a-priori unknown by testing for ignorance at the start followed by the acquisition of the requisite knowledge followed by the application of the knowledge on a completely novel problem. 4) FAIL: = your physics is wrong or your design is bad. PASS = design and physics are good. REPEAT THE ABOVE for all putative physics END when you get success...voila...the physics you dreamt up is the right one or as good as the right one. If the entity delivers the 'law of nature' then it has to have all the essential aspects of a visual experience needed for a successful scientific act. You can argue about the 'experience' within the entity afterwards...on a properly informed basis of real knowledge. Until then you're just waffling about theories. Such a test might involve reward through reverse-engineering chess. Initially chess ignorance is demonstrated...followed by repeated exposure to chess behaviour on a real board.followed by a demand to use chess behaviour in a completely environment and in a different manner...say to operate a machine that has nothing to do with chess but is metaphorically labelled to signal that chess rules apply to some aspect of its behaviour This proves that the laws underneath the external behaviour of the original chess pieces was internalised and abstracted...which contains all the essential ingredients of a scientific act on the unknown. You cannot do this without authentic connection to the distal external world of the chess pieces. You cannot train such an entity. The scientific act itself is the training. Neither testers nor tested can have any knowledge of the 'law of nature' or the environments to be encountered. A completely novel 'game' could be substituted for chess, for example. Any entity dependent on any sort of training will fail. You can't train for scientific outcomes. You can only build the necessities of scientific behaviour and then let it loose. You run this test on all putative theories of consciousness. If you can;'t build it you have no theory. If you build it and it fails, tough. If you build it and it passes your theory is right. You can't test for consciousness is a cultural catch phrase identical to man cannot fly. Just like the Wright Bros, we need to start to fly. Not pretend to fly. Or not fly and say we did Objective testing for consciousness is easy. Building the test and the entity...well that's not so easy but it is possible. A 'definition' of consciousness is irrelevant. Like every other circumstance in science...'laws' and physical phenomena that operate according to them are discovered, not defined. Humans did not wait for a definition of fire before cooking dinner with it. Why should consciousness be any different? cheers colin hales --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
--- On Wed, 11/12/08, Colin Hales [EMAIL PROTECTED] wrote: It is difficult but you can test for it objectively by demanding that an entity based on your 'theory of consciousness' deliver an authentic scientific act on the a-priori unknown using visual experience for scientific evidence. So a blind person is not conscious? -- Matt Mahoney, [EMAIL PROTECTED] --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
--- On Wed, 11/12/08, Harry Chesley [EMAIL PROTECTED] wrote: Matt Mahoney wrote: 2) It is real, as it clearly influences our thoughts. On the other hand, though it feels subjectively like it is qualitatively different from other aspects of the world, it probably isn't (but I'm open to being wrong here). The correct statement is that you believe it is real. Everybody does. Those who didn't, did not pass on their DNA. No, the correct statement is the one I made. It is real. We have empirical evidence that it is real since it influences observable actions. If you don't define consciousness in terms of an objective test, then you can say anything you want about it. -- Matt Mahoney, [EMAIL PROTECTED] --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
Matt Mahoney wrote: --- On Wed, 11/12/08, Colin Hales [EMAIL PROTECTED] wrote: It is difficult but you can test for it objectively by demanding that an entity based on your 'theory of consciousness' deliver an authentic scientific act on the a-priori unknown using visual experience for scientific evidence. So a blind person is not conscious? -- Matt Mahoney, [EMAIL PROTECTED] A blind person cannot behave scientifically in the manner of the sighted. The blind person cannot be a scientist 'of that which is visually evidenced'. As an objective test specifically for visual P-consciousness, the blind person's failure would prove the blind person has no visual P-consciousness. If a monkey passed the test then it would be proved visually P-conscious (as well as mighty smart!). A blind-sighted person would fail because they can't handle the radical novelty in the test. Again the test would prove they have no visual P-consciousness. A computer, if it passed, must have created inside itself all of the attributes of P-consciousness as utilised in vision applied to scientific evidence. You can argue about the details of any 'experience' only when armed with the physics _after_ the test is passed, when you can discuss the true nature of the physics involved from an authoritative position. If the requisite physics is missing the test subject will fail. That is the characteristic of a useful test. Unambiguous outcomes critically dependent on the presence of a claimed phenomenon. You don't even have to know the physics details. External behaviour is decisive and anyone could administer the test, provided it was set up properly. Note that experimental-scientists and applied scientists are literally scientific evidence of consciousness. They don't have to deliver anything except their normal science deliverables to complete the proof. They do nothing else but prove they are visually P-conscious for their entire lives. cheers, colin --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
Matt Mahoney wrote: If you don't define consciousness in terms of an objective test, then you can say anything you want about it. We don't entirely disagree about that. An objective test is absolutely crucial. I believe where we disagree is that I expect there to be such a test one day, while you claim there can never be. (I say don't /entirely/ agree because I think we can talk about things that are not completely defined -- in this case, I believe most people reading this do know the subjective feeling of consciousness and recognize that that's what I mean. A scientific exploration requires a more thorough definition, but we can still have some meaningful discourse without it, though we do risk running off into wildly unsubstantiated theories when we do.) --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
When people discuss the ethics of the treatment of artificial intelligent agents, it's almost always with the presumption that the key issue is the subjective level of suffering of the agent. This isn't the only possible consideration. One other consideration is our stance relative to that agent. Are we just acting in a selfish way, using the agent as simply a means to achieve our goals? I'll just leave that idea open as there are traditions that see value in de-emphasizing greed and personal acquisitiveness. Another consideration is the inherent value of self-determination. This is above any suffering that might be caused by being a completely controlled subject. One of the problems of slavery was just that it simply works better if you let people decide things for themselves. Similarly, just letting an artificial agent have autonomy for its own sake may just be a more effective thing than having it simply be a controlled subject. So I don't even think the consciousness of an artificial intelligent agent is completely necessary in considering the ethics of our stance towards it. We can consider our own emotional position and the inherent value of independence of thinking. andi --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
Mark Waser wrote: An understanding of what consciousness actually is, for starters. It is a belief. No it is not. And that statement (It is a belief) is a cop-out theory. An understanding of what consciousness is requires a consensus definition of what it is. For most people, it seems to be an undifferentiated mess that includes all of attentional components, intentional components, understanding components, and, frequently, experiential components (i.e. qualia). This mess was cleaned up a great deal when Chalmers took the simple step of dividing it into the 'easy' problems and the hard problem (which is the last one on your list). The easy problems do not have any philosophical depth to them; the hard problem seems to be a philosophical chasm. You are *very* correct to say that An 'understanding' of what consciousness is requires a consensus definition of what it is. My goal is to get a consensus definition, which then contains within it the explanation also. But, yes, if my explanation does not also include a definition that satisfies everyone as a good consensus definition, then it does not work. That is why Matt's it is a belief is not an explanation: it leaves so many questions unanswered that it will never make it as a consensus definition/explanation. We will see. My paper on the subject is almost finished. Richard Loosemore If you only buy into the first three and do it in a very concrete fashion, consciousness (and ethics) isn't all that tough. Or you can follow Alice and star debating the real meaning of the third and whether or not the truly fourth exists in anyone except yourself. Personally, if something has a will (intentionality/goals) that it can focus effectively (attentional and understanding), I figure that you'd better start treating it ethically for your own long-term self-interest. Of course, that then begs the question of what ethics is . . . . but I think that that is pretty easy to solve as well . . . . --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
--- On Tue, 11/11/08, Richard Loosemore [EMAIL PROTECTED] wrote: Would a program be conscious if it passes the Turing test? If not, what else is required? No. An understanding of what consciousness actually is, for starters. It is a belief. No it is not. And that statement (It is a belief) is a cop-out theory. No. Depending on your definition of consciousness, there is either an objective test for it or not. If consciousness results in an observable difference in behavior, then a machine that passes the Turing test must be conscious because there is no observable difference between it and a human. Or, if consciousness is not observable, then you must admit that the brain does something that cannot be explained by the known (computable) laws of physics. You conveniently avoid this inconsistency by refusing to define what you mean by consciousness. That is a cop-out. -- Matt Mahoney, [EMAIL PROTECTED] --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
Matt Mahoney wrote: --- On Tue, 11/11/08, Richard Loosemore [EMAIL PROTECTED] wrote: Would a program be conscious if it passes the Turing test? If not, what else is required? No. An understanding of what consciousness actually is, for starters. It is a belief. No it is not. And that statement (It is a belief) is a cop-out theory. No. Depending on your definition of consciousness, there is either an objective test for it or not. If consciousness results in an observable difference in behavior, then a machine that passes the Turing test must be conscious because there is no observable difference between it and a human. Or, if consciousness is not observable, then you must admit that the brain does something that cannot be explained by the known (computable) laws of physics. You conveniently avoid this inconsistency by refusing to define what you mean by consciousness. That is a cop-out. Your 'belief' explanation is a cop-out because it does not address any of the issues that need to be addressed for something to count as a definition or an explanation of the facts that need to be explained. My proposal is being written up now and will be available at the end of tomorrow. It does address all of the facts that need to be explained. Richard Loosemore --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
Matt Mahoney wrote: --- On Mon, 11/10/08, Richard Loosemore [EMAIL PROTECTED] wrote: Do you agree that there is no test to distinguish a conscious human from a philosophical zombie, thus no way to establish whether zombies exist? Disagree. What test would you use? A sophisticated assessment of the mechanisms inside the cognitive system. Would a program be conscious if it passes the Turing test? If not, what else is required? No. An understanding of what consciousness actually is, for starters. It is a belief. No it is not. And that statement (It is a belief) is a cop-out theory. Richard Loosemore --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
An understanding of what consciousness actually is, for starters. It is a belief. No it is not. And that statement (It is a belief) is a cop-out theory. An understanding of what consciousness is requires a consensus definition of what it is. For most people, it seems to be an undifferentiated mess that includes all of attentional components, intentional components, understanding components, and, frequently, experiential components (i.e. qualia). If you only buy into the first three and do it in a very concrete fashion, consciousness (and ethics) isn't all that tough. Or you can follow Alice and star debating the real meaning of the third and whether or not the truly fourth exists in anyone except yourself. Personally, if something has a will (intentionality/goals) that it can focus effectively (attentional and understanding), I figure that you'd better start treating it ethically for your own long-term self-interest. Of course, that then begs the question of what ethics is . . . . but I think that that is pretty easy to solve as well . . . . --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
--- On Tue, 11/11/08, Richard Loosemore [EMAIL PROTECTED] wrote: Your 'belief' explanation is a cop-out because it does not address any of the issues that need to be addressed for something to count as a definition or an explanation of the facts that need to be explained. As I explained, animals that have no concept of death have nevertheless evolved to fear most of the things that can kill them. Humans have learned to associate these things with death, and invented the concept of consciousness as the large set of features which distinguishes living humans from dead humans. Thus, humans fear the loss or destruction of consciousness, which is equivalent to death. Consciousness, free will, qualia, and good and bad are universal human beliefs. We should not confuse them with truth by asking the wrong questions. Thus, Turing sidestepped the question of can machines think? by asking instead can machines appear to think? Since we can't (by definition) distinguish doing something from appearing to do something, it makes no sense for us to make this distinction. Likewise, asking if it is ethical to inflict simulated pain on machines is asking the wrong question. Evolution favors the survival of tribes that practice altruism toward other tribe members and teach these ethical values to their children. This does not mean that certain practices are good or bad. If there was such a thing, then there would be no debate about war, abortion, euthanasia, capital punishment, or animal rights, because these questions could be answered experimentally. The question is not how should machines be treated? The question is how will we treat machines? My proposal is being written up now and will be available at the end of tomorrow. It does address all of the facts that need to be explained. I am looking forward to reading it. -- Matt Mahoney, [EMAIL PROTECTED] --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
This does not mean that certain practices are good or bad. If there was such a thing, then there would be no debate about war, abortion, euthanasia, capital punishment, or animal rights, because these questions could be answered experimentally. Given a goal and a context, there is absolutely such a thing as good or bad. The problem with the examples that you cited is that you're attempting to generalize to a universal answer across contexts (because I would argue that there is a useful universal goal) which is nonsensical. All of this can be answered both logically and experimentally if you just ask the right question instead of engaging in vacuous hand-waving about how tough it all is after you've mindlessly expanded your problem beyond solution. - Original Message - From: Matt Mahoney [EMAIL PROTECTED] To: agi@v2.listbox.com Sent: Tuesday, November 11, 2008 5:58 PM Subject: **SPAM** Re: [agi] Ethics of computer-based cognitive experimentation --- On Tue, 11/11/08, Richard Loosemore [EMAIL PROTECTED] wrote: Your 'belief' explanation is a cop-out because it does not address any of the issues that need to be addressed for something to count as a definition or an explanation of the facts that need to be explained. As I explained, animals that have no concept of death have nevertheless evolved to fear most of the things that can kill them. Humans have learned to associate these things with death, and invented the concept of consciousness as the large set of features which distinguishes living humans from dead humans. Thus, humans fear the loss or destruction of consciousness, which is equivalent to death. Consciousness, free will, qualia, and good and bad are universal human beliefs. We should not confuse them with truth by asking the wrong questions. Thus, Turing sidestepped the question of can machines think? by asking instead can machines appear to think? Since we can't (by definition) distinguish doing something from appearing to do something, it makes no sense for us to make this distinction. Likewise, asking if it is ethical to inflict simulated pain on machines is asking the wrong question. Evolution favors the survival of tribes that practice altruism toward other tribe members and teach these ethical values to their children. This does not mean that certain practices are good or bad. If there was such a thing, then there would be no debate about war, abortion, euthanasia, capital punishment, or animal rights, because these questions could be answered experimentally. The question is not how should machines be treated? The question is how will we treat machines? My proposal is being written up now and will be available at the end of tomorrow. It does address all of the facts that need to be explained. I am looking forward to reading it. -- Matt Mahoney, [EMAIL PROTECTED] --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?; Powered by Listbox: http://www.listbox.com --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
On Wed, Nov 12, 2008 at 8:58 AM, Matt Mahoney [EMAIL PROTECTED] wrote: As I explained, animals that have no concept of death have nevertheless evolved to fear most of the things that can kill them. Humans have learned to associate these things with death, and invented the concept of consciousness as the large set of features which distinguishes living humans from dead humans. Thus, humans fear the loss or destruction of consciousness, which is equivalent to death. So you're saying you're not a heavy drinker eh? Trent --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
Reality check *** Consciousness is an emergent spectrum of subjectivity spanning 600 mill. years of evolution involving mega-trillions of competing organisms, probably selecting for obscure quantum effects/efficiencies Our puny engineering/coding efforts could never approach this - not even in a million years. An outwardly pragmatic language simulation, however, is very do-able. John LaMuth www.forebrain.org www.emotionchip.net - Original Message - From: Richard Loosemore [EMAIL PROTECTED] To: agi@v2.listbox.com Sent: Tuesday, November 11, 2008 8:31 AM Subject: Re: [agi] Ethics of computer-based cognitive experimentation Mark Waser wrote: An understanding of what consciousness actually is, for starters. It is a belief. No it is not. And that statement (It is a belief) is a cop-out theory. An understanding of what consciousness is requires a consensus definition of what it is. For most people, it seems to be an undifferentiated mess that includes all of attentional components, intentional components, understanding components, and, frequently, experiential components (i.e. qualia). This mess was cleaned up a great deal when Chalmers took the simple step of dividing it into the 'easy' problems and the hard problem (which is the last one on your list). The easy problems do not have any philosophical depth to them; the hard problem seems to be a philosophical chasm. You are *very* correct to say that An 'understanding' of what consciousness is requires a consensus definition of what it is. My goal is to get a consensus definition, which then contains within it the explanation also. But, yes, if my explanation does not also include a definition that satisfies everyone as a good consensus definition, then it does not work. That is why Matt's it is a belief is not an explanation: it leaves so many questions unanswered that it will never make it as a consensus definition/explanation. We will see. My paper on the subject is almost finished. Richard Loosemore If you only buy into the first three and do it in a very concrete fashion, consciousness (and ethics) isn't all that tough. Or you can follow Alice and star debating the real meaning of the third and whether or not the truly fourth exists in anyone except yourself. Personally, if something has a will (intentionality/goals) that it can focus effectively (attentional and understanding), I figure that you'd better start treating it ethically for your own long-term self-interest. Of course, that then begs the question of what ethics is . . . . but I think that that is pretty easy to solve as well . . . . --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?; Powered by Listbox: http://www.listbox.com --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
--- On Tue, 11/11/08, Mark Waser [EMAIL PROTECTED] wrote: This does not mean that certain practices are good or bad. If there was such a thing, then there would be no debate about war, abortion, euthanasia, capital punishment, or animal rights, because these questions could be answered experimentally. Given a goal and a context, there is absolutely such a thing as good or bad. The problem with the examples that you cited is that you're attempting to generalize to a universal answer across contexts (because I would argue that there is a useful universal goal) which is nonsensical. All of this can be answered both logically and experimentally if you just ask the right question instead of engaging in vacuous hand-waving about how tough it all is after you've mindlessly expanded your problem beyond solution. That's what I just said. You have to ask the right questions. -- Matt Mahoney, [EMAIL PROTECTED] --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
Matt Mahoney wrote: --- On Tue, 11/11/08, Richard Loosemore [EMAIL PROTECTED] wrote: Your 'belief' explanation is a cop-out because it does not address any of the issues that need to be addressed for something to count as a definition or an explanation of the facts that need to be explained. As I explained, animals that have no concept of death have nevertheless evolved to fear most of the things that can kill them. Humans have learned to associate these things with death, and invented the concept of consciousness as the large set of features which distinguishes living humans from dead humans. Thus, humans fear the loss or destruction of consciousness, which is equivalent to death. Consciousness, free will, qualia, and good and bad are universal human beliefs. We should not confuse them with truth by asking the wrong questions. Thus, Turing sidestepped the question of can machines think? by asking instead can machines appear to think? Since we can't (by definition) distinguish doing something from appearing to do something, it makes no sense for us to make this distinction. The above two paragraphs STILL do not address any of the issues that need to be addressed for something to count as a definition, or an explanation of the facts that need to be explained. Richard Loosemore --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
John LaMuth wrote: Reality check *** Consciousness is an emergent spectrum of subjectivity spanning 600 mill. years of evolution involving mega-trillions of competing organisms, probably selecting for obscure quantum effects/efficiencies Our puny engineering/coding efforts could never approach this - not even in a million years. An outwardly pragmatic language simulation, however, is very do-able. John LaMuth It is not. And we can. Richard Loosemore --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
[EMAIL PROTECTED] wrote: When people discuss the ethics of the treatment of artificial intelligent agents, it's almost always with the presumption that the key issue is the subjective level of suffering of the agent. This isn't the only possible consideration. One other consideration is our stance relative to that agent. Are we just acting in a selfish way, using the agent as simply a means to achieve our goals? I'll just leave that idea open as there are traditions that see value in de-emphasizing greed and personal acquisitiveness. Another consideration is the inherent value of self-determination. This is above any suffering that might be caused by being a completely controlled subject. One of the problems of slavery was just that it simply works better if you let people decide things for themselves. Similarly, just letting an artificial agent have autonomy for its own sake may just be a more effective thing than having it simply be a controlled subject. So I don't even think the consciousness of an artificial intelligent agent is completely necessary in considering the ethics of our stance towards it. We can consider our own emotional position and the inherent value of independence of thinking. andi I'm inclined to agree - this will be an issue in the future... if you have a robot helper and someone comes by and beats it to death in front of your kids, who have some kind of attachment to it...a relationship... then crime (i) may be said to be the psychological damage to the children. Crime (ii) is then the murder and whatever one knows of suffering inflicted on the robot helper. Ethicists are gonna have all manner of novelty to play with. cheers colin --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
RE: [agi] Ethics of computer-based cognitive experimentation
From: Richard Loosemore [mailto:[EMAIL PROTECTED] John LaMuth wrote: Reality check *** Consciousness is an emergent spectrum of subjectivity spanning 600 mill. years of evolution involving mega-trillions of competing organisms, probably selecting for obscure quantum effects/efficiencies Our puny engineering/coding efforts could never approach this - not even in a million years. An outwardly pragmatic language simulation, however, is very do-able. John LaMuth It is not. And we can. I thought what he said was a good description more or less. Out of 600 millions years there may be only a fraction of that which is an improvement but it's still there. How do you know, beyond a reasonable doubt, that any other being is conscious? At some point you have to trust that others are conscious, in the same species, you bring them into your recursive loop of consciousness component mix. A primary component of consciousness is a self definition. Conscious experience is unique to the possessor. It is more than a belief that the possessor herself is conscious but others who appear conscious may be just that, appearing to be conscious. Though at some point there is enough feedback between individuals and/or a group to share consciousness experience. Still though, is it really necessary for an AGI to be conscious? Except for delivering warm fuzzies to the creators? Doesn't that complicate things? Shouldn't the machines/computers be slaves to man? Or will they be equal/superior. It's a dog-eat-dog world out there. I just want things to be taken care of and no issues. Consciousness brings issues. Intelligence and consciousness are separate. John --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=120640061-aded06 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
Matt Mahoney wrote: --- On Fri, 11/7/08, Richard Loosemore [EMAIL PROTECTED] wrote: The question of whether a test is possible at all depends on the fact that there is a coherent theory behind the idea of consciousness. Would you agree that consciousness is determined by a large set of attributes that are present in living human brains but absent in dead human brains? Yes Do you agree that there is no test to distinguish a conscious human from a philosophical zombie, thus no way to establish whether zombies exist? Disagree. Would a program be conscious if it passes the Turing test? If not, what else is required? No. An understanding of what consciousness actually is, for starters. Richard Loosemore -- Matt Mahoney, [EMAIL PROTECTED] --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?; Powered by Listbox: http://www.listbox.com --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=117534816-b15a34 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
--- On Mon, 11/10/08, Richard Loosemore [EMAIL PROTECTED] wrote: Do you agree that there is no test to distinguish a conscious human from a philosophical zombie, thus no way to establish whether zombies exist? Disagree. What test would you use? Would a program be conscious if it passes the Turing test? If not, what else is required? No. An understanding of what consciousness actually is, for starters. It is a belief. -- Matt Mahoney, [EMAIL PROTECTED] --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=117534816-b15a34 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
Matt Mahoney wrote: --- On Mon, 11/10/08, Richard Loosemore [EMAIL PROTECTED] wrote: Do you agree that there is no test to distinguish a conscious human from a philosophical zombie, thus no way to establish whether zombies exist? Disagree. What test would you use? The test will be published in the next couple of months in the Open AI journal. = An objective test for scientific behaviour. I call it the 'PCST' for P-Conscious Scientist Test. You can't be a scientist without being visually P-conscious to experience your evidence. You can't deny the test without declaring scientists devoid of consciousness whilst demanding it be used for all scientific evidence in a verifiable way AND whilst investing in an entire science paradigm Neural Correlates of Consciousness dedicated to scientific exploration of P-consciousnessThe logic's pretty good and it's easy to design an objective test demanding delivery of a 'law of nature'. The execution, however, is logistically difficult++. BUT...At least it's doable. A hard test is better than no test at all, which is what we currently have. When it comes out I'll let you know. RE ETHICS..I say this in the paper: As was recognised by Gamez [35], one cannot help but notice that there is also a secondary ethical 'bootstrap' process. Once a single subject passes the PCST, for the first time ever in certain circumstances there will be a valid scientific reason obliging all scientists to consider the internal life of an artefact as potentially having some level of equivalence to that of a laboratory animal, possibly deserving of similar ethical treatment. Until that event occurs, however, all such discussions are best considered moot. cheers colin --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=117534816-b15a34 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
Matt Mahoney wrote: --- On Wed, 11/5/08, Richard Loosemore [EMAIL PROTECTED] wrote: In the future (perhaps the near future) it will be possible to create systems that will have their own consciousness. *Appear* to have consciousness, or do you have a test? Yes. But the test depends on an understanding of the system architecture. The question of whether a test is possible at all depends on the fact that there is a coherent theory behind the idea of consciousness. Stepping back for the moment, the entire question of ethics depends crucially on your theory of how consciousness arises. We talk about such things as if we can answer the question of why it is OK to stomp on a roach but not a puppy by studying the brains of roaches and puppies. It is not possible to look at the brains and decide whether or not is okay to stomp, but we can decide whether or not the brain has a significant level of consciousness that it is comparable to ours. That is vital information in making a reasoned judgement of stompworthiness. For the record, I am treading carefully. As far as what happens in my lab, I will explicitly put in place measures to ensure that AGI systems that do have a chance of reasonably high levels of consciousness will have the fullest possible ethical protections. I cannot speak for anyone else, but that is my policy. Now I am curious. Given a program P, what is your lab's criteria for determining whether P is conscious? Complicated. I'll get right backtya on that ;-) Richard Loosemore --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=117534816-b15a34 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
--- On Fri, 11/7/08, Richard Loosemore [EMAIL PROTECTED] wrote: The question of whether a test is possible at all depends on the fact that there is a coherent theory behind the idea of consciousness. Would you agree that consciousness is determined by a large set of attributes that are present in living human brains but absent in dead human brains? Do you agree that there is no test to distinguish a conscious human from a philosophical zombie, thus no way to establish whether zombies exist? Would a program be conscious if it passes the Turing test? If not, what else is required? -- Matt Mahoney, [EMAIL PROTECTED] --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=117534816-b15a34 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
On Thu, Nov 6, 2008 at 12:55 AM, Harry Chesley [EMAIL PROTECTED] wrote: Personally, I'm not making an AGI that has emotions... So you take the view that, despite our minimal understanding of the basis of emotions, they will only arise if designed in, never spontaneously as an emergent property? So you can safely ignore the ethics question. Well, my AGI system would take special measures to ensure that emotions do *not* emerge, by making the system acquire *knowledge* of human values instead of having emotions occurring at the AGI's *perceptual* level. YKY --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=117534816-b15a34 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
On Wed, Nov 5, 2008 at 7:35 AM, Matt Mahoney [EMAIL PROTECTED] wrote: Personally, I'm not making an AGI that has emotions, and I doubt if emotions are generally desirable in AGIs, except when the goal is to make human companions (and I wonder why people need them anyway, given that there're so many -- *too* many -- human beings around already). People may want to simulate loved ones who have died, if the simulation is accurate enough to be indistinguishable. People may also want to simulate themselves in the same way, in the belief it will make them immortal. Yeah, I should qualify my statement: different people will want different things out of AGI technology. Some want brain emulation of themselves or loved ones, some want android companions, etc. All these things take up free energy (a scarce resource on earth), so it is just a new form of the overpopulation problem. I am not against any particular form of AGI application; I just want to point out that AGI-with-emotions is not necessary goal of AGI. --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=117534816-b15a34 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
YKY:I just want to point out that AGI-with-emotions is not necessary goal of AGI. Which AGI as distinct from narrow AI problems do *not* involve *incalculable and possibly unmanageable risks*? - a)risks that the process of problem-solving will be interminable? b)risks that the agent does not have the skills necessary for the problem's solution? c)risks that the agent hasn't defined the problem properly? That's what the emotion of fear is - (one of the emotions essential for AGI) - a system alert to incalculable and possibly unmanageable risks. That's what the classic fight-or-flight response entails - maybe I can deal with this danger but maybe I can't and better avoid it fast. --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=117534816-b15a34 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
YKY, As I was saying, before I so rudely interrupted myself - re the narrow AI vs AGI problem difference: *the syllogistic problems of logic - is Aristotle mortal? etc - which you mainly use as examples - are narrow AI problems, which can be solved according to precise rules however: *metacognitive problems - like *which logic should I use for syllogistic problems, eg PLN/NARS? - (which also concerns you) - are AGI problems; there are no rules for solving them, and no definitive solutions, only possible, temporary resolutions to someone's satisfaction. Those are problems which you have been discussing and could continue to discuss interminably. And they are also problems which you will have - and any agent considering, should have - fear considering, because you can get endlessly bogged down in them [n.b. psychologically, fear comes in many different degrees from panic to mild wariness] similarly *is cybersex sex? (another of your problems) - if treated by some artificial logic with artificial rules, (which might end up saying yes, approx. 0.60 % sex), is a narrow AI problem; however, if treated realistically, *philosophically*, relying on language, this is an AGI problem, which can be and may well be considered interminably by real philosophers (and lawyers) into the next century, (*did* Clinton have sex?) and for which there are neither definitive rules nor solution . Again fear is, and has to be a part of considering such problems - how much life do you have to spend on them? Even the biggest computer brain in the world, the superestAGI will not be able to solve them definitively, and must be afraid of them ditto: *Any philosophical problem of definition: what is mind? What is consciousness? What is intelligence? Again these are infinitely open-ended, open-means problems, which have atttracted and will continue to attract interminable consideration. You are, and should be, afraid, of getting too deep into them *Any linguistic problem of definition: what does honour,beautiful, big small etc mean? is an AGI problem AFAIK literally any word in the language is open to endless definition and redefinition and essentially an AGI problem. By contrast, *what is ETFUBAIL an anagram of? is a narrow AI problem - and no need for any fear there. *Defining/describing almost anything - describe YKY or Ben Goertzel; what kind of guys/ programmers are they? - are AGI problems. You could consider them forever. You may be skilled at resolving them quickly, and able to come up with a brief description, but that again while perhaps satisfactory will never do the subject even remotely perfect justice, and could be endlessly improved and sophisticated. In general, your instinct - and most AGI-ers' instinct - seems to be, whenever confronted with an AGI problem, to try and reduce it to a narrow AGI problem - from a real, open-ended/ open-means-and-rules to an artificial, closed-ended, closed-means-and-rules problem. Then, yes, you don't need fear and other emotions, but that's not AGI. YKY:I just want to point out that AGI-with-emotions is not necessary goal of AGI. Which AGI as distinct from narrow AI problems do *not* involve *incalculable and possibly unmanageable risks*? - a)risks that the process of problem-solving will be interminable? b)risks that the agent does not have the skills necessary for the problem's solution? c)risks that the agent hasn't defined the problem properly? That's what the emotion of fear is - (one of the emotions essential for AGI) - a system alert to incalculable and possibly unmanageable risks. That's what the classic fight-or-flight response entails - maybe I can deal with this danger but maybe I can't and better avoid it fast. --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?; Powered by Listbox: http://www.listbox.com --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=117534816-b15a34 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
On 11/4/2008 2:53 PM, YKY (Yan King Yin) wrote: Personally, I'm not making an AGI that has emotions... So you take the view that, despite our minimal understanding of the basis of emotions, they will only arise if designed in, never spontaneously as an emergent property? So you can safely ignore the ethics question. --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=117534816-b15a34 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
On 11/4/2008 3:31 PM, Matt Mahoney wrote: To answer your (modified) question, consciousness is detected by the activation of a large number of features associated with living humans. The more of these features are activated, the greater the tendency to apply ethical guidelines to the target that we would normally apply to humans. For example, monkeys are more like humans than mice, which are more like humans than insects, which are more like humans than programs. It does not depend on a single feature. If I understand correctly, you're saying that there is no such thing as objective ethics, and that our subjective ethics depend on how much we identify/empathize with another creature. I grant this as a possibility, in which case I guess my question should be viewed as subjective. I.e., how do I tell when something is sufficiently close to me, without being able to see all the features directly, that I need to worry about the ethics subjectively? Let me give an example: If I take a person and put them in a box, so that I can see none of their features or know how similar they are to me, I still consider it unethical to conduct certain experiments on them. This is because I believe those important similar features are there, I just can't see them. Similarly, I believe at some point in AGI development, features similar to my own mind will arise, but since they will be obscured by a very different (and incomplete) implementation from my own, they may not be obvious, even though I believe they are there. So although you've changed the phrasing of the question to a degree, the question remains. (Note: You could argue that ethics, being subjective, are irrelevant, and while that may be true, I'm too squeamish to take that view, which also leads to allowing arbitrary experiments on people.) --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=117534816-b15a34 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
--- On Wed, 11/5/08, YKY (Yan King Yin) [EMAIL PROTECTED] wrote: On Wed, Nov 5, 2008 at 7:35 AM, Matt Mahoney [EMAIL PROTECTED] wrote: Personally, I'm not making an AGI that has emotions, and I doubt if emotions are generally desirable in AGIs, except when the goal is to make human companions (and I wonder why people need them anyway, given that there're so many -- *too* many -- human beings around already). People may want to simulate loved ones who have died, if the simulation is accurate enough to be indistinguishable. People may also want to simulate themselves in the same way, in the belief it will make them immortal. Yeah, I should qualify my statement: different people will want different things out of AGI technology. Some want brain emulation of themselves or loved ones, some want android companions, etc. All these things take up free energy (a scarce resource on earth), so it is just a new form of the overpopulation problem. I am not against any particular form of AGI application; I just want to point out that AGI-with-emotions is not necessary goal of AGI. I agree. My own AGI design does not require emotion, assuming the goal is to automate the economy. My proposed solution is a decentralized message passing network that implements distributed compression of the world's knowledge by trading in an economy where information has negative value. Peers mutually benefit by trading messages that are hard to compress by the sender and easy to compress by the receiver. This has the effect that peers tend to specialize and that messages get routed to the right experts. If our language model is a simple unigram word model, then we have a distributed implementation of Salton's tf-idf information retrieval model. A language model uses three types of learning: eidetic (short term) memory, association of concepts (e.g. words) in eidetic memory, and learning new concepts by clustering in context space. Vision is learned the same way. In both cases, reinforcement learning (a prerequisite of emotion) is not required. If the goal of AGI is uploading or simulating humans, then of course it is necessary to simulate human emotions. Also if we allow agents to modify themselves and reproduce, then evolution will favor emotions such as fear of death, greed, tribal altruism, and the desire to reproduce. -- Matt Mahoney, [EMAIL PROTECTED] --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=117534816-b15a34 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
--- On Wed, 11/5/08, Harry Chesley [EMAIL PROTECTED] wrote: If I understand correctly, you're saying that there is no such thing as objective ethics, and that our subjective ethics depend on how much we identify/empathize with another creature. I grant this as a possibility, in which case I guess my question should be viewed as subjective. I.e., how do I tell when something is sufficiently close to me, without being able to see all the features directly, that I need to worry about the ethics subjectively? Let me give an example: If I take a person and put them in a box, so that I can see none of their features or know how similar they are to me, I still consider it unethical to conduct certain experiments on them. This is because I believe those important similar features are there, I just can't see them. It is surprisingly easy for humans to lessen their anxiety by blocking the stimuli that makes another suffering person seem human. An important feature of the Milgram experiments ( http://en.wikipedia.org/wiki/Milgram_experiment ) was that the torturer could not see the victim. Likewise, people who wouldn't hesitate to jump in the water to save a drowning child will do nothing to stop the suffering of millions of starving refugees on the other side of the world. I don't mean to imply that we should behave differently. I am just describing how the human ethical belief model works. -- Matt Mahoney, [EMAIL PROTECTED] --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=117534816-b15a34 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
Harry Chesley wrote: On 11/4/2008 3:31 PM, Matt Mahoney wrote: To answer your (modified) question, consciousness is detected by the activation of a large number of features associated with living humans. The more of these features are activated, the greater the tendency to apply ethical guidelines to the target that we would normally apply to humans. For example, monkeys are more like humans than mice, which are more like humans than insects, which are more like humans than programs. It does not depend on a single feature. If I understand correctly, you're saying that there is no such thing as objective ethics, and that our subjective ethics depend on how much we identify/empathize with another creature. I grant this as a possibility, in which case I guess my question should be viewed as subjective. I.e., how do I tell when something is sufficiently close to me, without being able to see all the features directly, that I need to worry about the ethics subjectively? Let me give an example: If I take a person and put them in a box, so that I can see none of their features or know how similar they are to me, I still consider it unethical to conduct certain experiments on them. This is because I believe those important similar features are there, I just can't see them. Similarly, I believe at some point in AGI development, features similar to my own mind will arise, but since they will be obscured by a very different (and incomplete) implementation from my own, they may not be obvious, even though I believe they are there. So although you've changed the phrasing of the question to a degree, the question remains. (Note: You could argue that ethics, being subjective, are irrelevant, and while that may be true, I'm too squeamish to take that view, which also leads to allowing arbitrary experiments on people.) I can answer your questions about ethics from the perspective of someone trying to build real AGI systems that are similar to human minds. In principle, there is no reason why an AGI system should not be in need of ethical protection, but it depends on the system. At the moment, the design of AGI systems is such that there is no immediate danger of an intelligence being created that is sufficiently self-aware that it would have anything resembling human consciousness. Simply put, present systems are almost certainly not capable of feeling pain or needing ethical protection. This statement would require quite a lengthy justification, but I think it is a fairly safe conclusion. In the future (perhaps the near future) it will be possible to create systems that will have their own consciousness. However, even then there will be quite drastic differences between different designs, and we will have to proceed quite carefully. For example, it will be possible to create systems that are fundamentally designed to want to do certain things, like serving humans, or like living in virtual worlds where they do not have contact with the real world. Those systems should not be viewed as 'enslaved because, in point of fact, they would want to do what they do: their behavior is what makes them happy, and liberating them from this behavior would make them unhappy. It would not be ethical to take such a system and treat it as if it were a human slave that needed to be liberated. This would never be true for any human being (no human being truly would be happy as a slave), but it would be fundamentally true in the case of this hypothetical AGI system. This possibility of creating systems that get fulfilment in ways that are different from the ways that humans get fulfilment must be taken into account when ethical considerations are evaluated. Stepping back for the moment, the entire question of ethics depends crucially on your theory of how consciousness arises. There is no consensus on this at the moment, but it is important to understand that any judgement about ethics, either way, can only be made in the context of a statement about what exactly the theory of consciousness is that lies behind the statement. Nobody could simply say, for example, Let's assume that all AI systems need ethical protection right now, as a default assumption, because that kind of default has an *implicit* theory of consciousness behind it that is pure guesswork, and is not supported by anything we understand about consciousness at the moment. For the record, I am treading carefully. As far as what happens in my lab, I will explicitly put in place measures to ensure that AGI systems that do have a chance of reasonably high levels of consciousness will have the fullest possible ethical protections. I cannot speak for anyone else, but that is my policy. Richard Loosemore --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify
Re: [agi] Ethics of computer-based cognitive experimentation
--- On Wed, 11/5/08, Richard Loosemore [EMAIL PROTECTED] wrote: In the future (perhaps the near future) it will be possible to create systems that will have their own consciousness. *Appear* to have consciousness, or do you have a test? Stepping back for the moment, the entire question of ethics depends crucially on your theory of how consciousness arises. We talk about such things as if we can answer the question of why it is OK to stomp on a roach but not a puppy by studying the brains of roaches and puppies. For the record, I am treading carefully. As far as what happens in my lab, I will explicitly put in place measures to ensure that AGI systems that do have a chance of reasonably high levels of consciousness will have the fullest possible ethical protections. I cannot speak for anyone else, but that is my policy. Now I am curious. Given a program P, what is your lab's criteria for determining whether P is conscious? -- Matt Mahoney, [EMAIL PROTECTED] --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=117534816-b15a34 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
On Wed, Nov 5, 2008 at 6:05 AM, Harry Chesley [EMAIL PROTECTED] wrote: The question of when it's ethical to do AGI experiments has bothered me for a while. It's something that every AGI creator has to deal with sooner or later if you believe you're actually going to create real intelligence that might be conscious. The following link is a blog essay on the subject, which describes my current thinking on the subject, such as it is. There's clearly much more that needs to be worked out. Comments, either here or at the blog, would be appreciated. http://www.mememotes.com/meme_motes/2008/11/ethical-experimentation-on-cognitive-entities.html Personally, I'm not making an AGI that has emotions, and I doubt if emotions are generally desirable in AGIs, except when the goal is to make human companions (and I wonder why people need them anyway, given that there're so many -- *too* many -- human beings around already). YKY --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=117534816-b15a34 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
--- On Tue, 11/4/08, Harry Chesley [EMAIL PROTECTED] wrote: The question of when it's ethical to do AGI experiments has bothered me for a while. That's because you're asking the wrong question. Don't confuse belief with truth. The question is: what ethical guidelines will we (not should we) adopt regarding experiments with AGI? Some background: all animals with nervous systems advanced enough to be capable of reinforcement learning have evolved fears of most of the things that can kill them. Humans (and possibly other advanced animals) have learned the concept of death and developed the concept of consciousness, defined as the set of features that distinguishes alive from dead, for example, ability to communicate, to move, to think, to remember things, to experience things, to make decisions, to feel emotions, and so on. The list is quite large. In short, consciousness is the set of attributes that we fear losing. Furthermore, humans are social animals. We evolved (both genetically and memetically through culture) an ethical system that respects consciousness in other members of our tribe. Fear, consciousness, qualia, free will, and good and bad, are all beliefs. These concepts are useful for describing human behavior. It is not necessary to assume that any of these things actually exist in order to do so. To answer your (modified) question, consciousness is detected by the activation of a large number of features associated with living humans. The more of these features are activated, the greater the tendency to apply ethical guidelines to the target that we would normally apply to humans. For example, monkeys are more like humans than mice, which are more like humans than insects, which are more like humans than programs. It does not depend on a single feature. For example, the program http://www.mattmahoney.net/autobliss.txt simulates reinforcement learning in a simple agent. You can run it with the second and third argument both negative, meaning it is punished no matter what it does. You might consider such an experiment unethical if performed on monkeys but not on insects, and certainly not on this program. As a second example, the video game Grand Theft Auto allows you to have simulated sex with prostitutes and then beat them to death to get your money back. While playing, I declined to do so, even though it was irrational with respect to the goal of attaining the highest possible score. -- Matt Mahoney, [EMAIL PROTECTED] --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=117534816-b15a34 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
--- On Tue, 11/4/08, YKY (Yan King Yin) [EMAIL PROTECTED] wrote: Personally, I'm not making an AGI that has emotions, and I doubt if emotions are generally desirable in AGIs, except when the goal is to make human companions (and I wonder why people need them anyway, given that there're so many -- *too* many -- human beings around already). People may want to simulate loved ones who have died, if the simulation is accurate enough to be indistinguishable. People may also want to simulate themselves in the same way, in the belief it will make them immortal. -- Matt Mahoney, [EMAIL PROTECTED] --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=117534816-b15a34 Powered by Listbox: http://www.listbox.com
Re: [agi] Ethics of computer-based cognitive experimentation
--- On Tue, 11/4/08, Trent Waddington [EMAIL PROTECTED] wrote: On Wed, Nov 5, 2008 at 9:31 AM, Matt Mahoney [EMAIL PROTECTED] wrote: As a second example, the video game Grand Theft Auto allows you to have simulated sex with prostitutes and then beat them to death to get your money back. While playing, I declined to do so, even though it was irrational with respect to the goal of attaining the highest possible score. Good for you. You have principles, you stuck by them, even when it meant depriving yourself of something (a trivial something, but something). Remember good is only a belief. I behave the way I am programmed to. My only fear is that people like you often turn into people who want the game banned so that no-one else may engage in the activity you disagree with - and this is why discussions about ethical treatment of AGI systems makes me gag... because inevitably someone is going to say there oughta be a law and the entire industry will come to a screeching halt. Don't say it won't happen.. remember the blanket ban on cloning technology. The issue here is people are concerned about teaching criminal behavior to children. (Again, I don't claim we should or shouldn't). So far there is no concern about the treatment of programs. Our laws regarding animal treatment are similar. We don't object so much to the suffering of animals (e.g. raising chickens in tiny cages for slaughter) as we do to public displays of it (e.g. cock fighting). AGI researchers could adopt a similar approach, i.e. not talking about their programs in human terms. But eventually, we will have to confront the issue. As I posted earlier, people will want to simulate their deceased loved ones, and once the technology is demonstrated, themselves. The first uploads are likely to be rough: no embodiment, a lot of incomplete and made-up memories, and poorly done AI, barely passing the Turing test. As technology improves (surveillance, computing power, better AI algorithms, perhaps brain scanning), the uploads will get more realistic. The problem* is when we give uploads legal and property rights. Humans have an incentive to do so, not just out of ethical concerns, but also for selfish reasons; first to alleviate grief by simulating loved ones, and second when people pass their rights to their simulations after they die in the belief that doing so will make them immortal. *I don't mean to imply that human extinction, or viewed another way, our evolution into a non-DNA based life form, is good or bad. However, it is normal for people to make such judgments. -- Matt Mahoney, [EMAIL PROTECTED] --- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244id_secret=117534816-b15a34 Powered by Listbox: http://www.listbox.com