gin001@cdc835.cdc.polimi.it (Mauro Cicognini) (05/16/91)
It may be a naive idea, but my assertion is that artificial intelligence is not worth for any economical results it may produce. In fact, it will always be less expensive to train a human being to do a certain intel- lectual job than to devise an artificial system to do the same thing. This because the machine has first to be imbued with "artificial" intel- ligence; and this task has yet to prove itself easy. I doubt that it ever will, though this is only a personal opinion. My point is that, even if we reach the point where we can "make" intellig- ence out of silicon (or gallium arsenide), it will always be more costly than to have intelligence produced to old way, that is, to make babies. Than why all this fuss about AI? I think this is a just and good effort, because of the enormous consequences it has already had and it'll have in the future on the way we understand ourselves. I mean that the research on AI has stimulated research on intelligence at large, and so including the way we memorize things, the way we understand, we general- ize, and so on. Let's not forget the studies on artificial vision and hearing, manipulation, et cetera. We have this way taken away this subject from the exclusive domain of philosophers and literature-major people, and merged it with the main- stream of scientifical research; paving the way for a countless number of applications in a world that relies ever more heavily on machines. I believe that the one most important consequence af AI will be to ease our survival in such a world, improving our relations with ourselves, other humans and, last but not least, the machines we cope with.
ziane@nuri.inria.fr (ziane mikal @) (05/21/91)
In article <1991May16.134123.6335@cdc835.cdc.polimi.it> gin001@cdc835.cdc.polimi.it (Mauro Cicognini) writes: >It may be a naive idea, but my assertion is that artificial intelligence >is not worth for any economical results it may produce. In fact, it will >always be less expensive to train a human being to do a certain intel- >lectual job than to devise an artificial system to do the same thing. >This because the machine has first to be imbued with "artificial" intel- >ligence; and this task has yet to prove itself easy. I doubt that it ever >will, though this is only a personal opinion. > >My point is that, even if we reach the point where we can "make" intellig- >ence out of silicon (or gallium arsenide), it will always be more costly >than to have intelligence produced to old way, that is, to make babies. Ridiculous ! First how can you forecast this cost so confidently ? Second, computers may be able to do things that human beings cannot or don't like. Computers may be able to do those things faster and better.. Third, you may duplicate computer knowledge much more easily than human knowledge ! etc, etc, etc... Mikal.
TKUSTER@auvm.auvm.edu (Todd) (05/21/91)
In article <1991May16.134123.6335@cdc835.cdc.polimi.it>, gin001@cdc835.cdc.polimi.it (Mauro Cicognini) says: > >It may be a naive idea, but my assertion is that artificial intelligence >is not worth for any economical results it may produce. In fact, it will >always be less expensive to train a human being to do a certain intel- >lectual job than to devise an artificial system to do the same thing. >This because the machine has first to be imbued with "artificial" intel- >ligence; and this task has yet to prove itself easy. I doubt that it ever >will, though this is only a personal opinion. > >My point is that, even if we reach the point where we can "make" intellig- >ence out of silicon (or gallium arsenide), it will always be more costly >than to have intelligence produced to old way, that is, to make babies. > The main reason it's still worth researching AI is because while it currently cheaper to produce a human to do the thinking in the future it will become cheaper to have the machine do the thinking. You also have to consider the cost and time from pre-birth to the time you have a thinking person capable of doing an intellectual job. Now consider once you devise an AI system that can do the job they can be mass produced at a fraction of the cost and time. Just a thought...
cam@aifh.ed.ac.uk (Chris Malcolm) (05/27/91)
In article <1991May16.134123.6335@cdc835.cdc.polimi.it> gin001@cdc835.cdc.polimi.it (Mauro Cicognini) writes: >It may be a naive idea, but my assertion is that artificial intelligence >is not worth for any economical results it may produce. In fact, it will >always be less expensive to train a human being to do a certain intel- >lectual job than to devise an artificial system to do the same thing. >This because the machine has first to be imbued with "artificial" intel- >ligence; and this task has yet to prove itself easy. I doubt that it ever >will, though this is only a personal opinion. >My point is that, even if we reach the point where we can "make" intellig- >ence out of silicon (or gallium arsenide), it will always be more costly >than to have intelligence produced to old way, that is, to make babies. You're a decade or so out of date with this argument. 1. A chess machine which can beat me. I can buy one of these for approx $100. Training a human being to do this is a LOT more expensive. In fact, since most people couldn't be trained to this level of performance, actually selecting the individual, before starting the training, would probably cost more than buying the machine. 2. A general purpose mathematics assistant, which can do linear algebra, find roots, integrate, draw graphs, etc., can be bought (in the form of a computer program) for a good deal less money than training a human being to the appropriate level; and once again, you would first have to select a suitable trainee. 3. A language translation hacker. While machines which can produce high quality translations between natural human languages do not yet exist, there already exist machines which can translate badly, but well enough that they are in regular use by professional document translators. The machine produces a rough draft, and the human corrects it with an editor, enabling the human to do a great deal more translation in a given time. Once again, training a human to the same level of expertise would be a lot more expensive than buying the program. In all these cases not only are the machines VERY much cheaper to buy than training an appropriate human, but they are also VERY much cheaper to run than the salary the trained human would require. Of course, there is one sense in which you are correct: the man-years of research that were necessary before it was possible to build one of these machines did far exceed the cost of training ONE suitable human candidate for the job, just as the man-years which went into inventing and building the telephone network far exceed the cost of you having to write letters instead of phoning for all of your life. Fortunately for industry that is not the cost-equation by which research and development is costed, and fortunately for you there are so many people who want telephones that each subscriber need bear only a minute and easily affordable fraction of the cost of the world's telephone network. -- Chris Malcolm cam@uk.ac.ed.aipna +44 (0)31 667 1011 x2550 Department of Artificial Intelligence, Edinburgh University 5 Forrest Hill, Edinburgh, EH1 2QL, UK DoD #205
jacob@latcs1.lat.oz.au (Jacob L. Cybulski) (05/28/91)
In article <1991May16.134123.6335@cdc835.cdc.polimi.it> gin001@cdc835.cdc.polimi.it (Mauro Cicognini) writes: > It may be a naive idea, but my assertion is that artificial intelligence > is not worth for any economical results it may produce. In fact, it will > always be less expensive to train a human being to do a certain intel- > lectual job than to devise an artificial system to do the same thing. Yes, you are right! Constructing an intelligent artificial system is usually a very expensive tasks. It is considerably more expensive than training an individual to perform the same task effectively. In some cases, devising the process of designing and implementing a sophisticated expert system may take 100s or even 1000s of man-years as compared with 10s of years of expert education and experience. Where you are missing the point is that an intelligent system may represent the knowledge of many experts that is too voluminous and complex for a single person to master. Secondly that, once the system is constructed, the replication of all the expert knowledge and skill is very quick and cheap. So we may be looking at several millions in development and productisation of an AI package which represents the knowledge of 100s of experts. But their collective education and expertise must also have costed equal the amount. The resulting product, however, could complement non-experts in their work at a very nominal cost. Considering an average mainframe Expert System cost at $100,000, and its use by 10 persons only, this would result in roughly $10,000 per person. $10K does not buy a lot of experts (even at the state education prices)! Jacob L. Cybulski Amdahl Australian Intelligent Tools Programme Department of Computer Science and Computer Engineering La Trobe University Bundoora, Vic 3083, Australia Phone: +613 479 1270 Fax: +613 470 4915 Telex: AA 33143 EMail: jacob@latcs1.oz.au
markv@pixar.com (Mark VandeWettering) (06/13/91)
>It may be a naive idea, but my assertion is that artificial intelligence >is not worth for any economical results it may produce. In fact, it will >always be less expensive to train a human being to do a certain intel- >lectual job than to devise an artificial system to do the same thing. >This because the machine has first to be imbued with "artificial" intel- >ligence; and this task has yet to prove itself easy. I doubt that it ever >will, though this is only a personal opinion. Similar thinking would be along the lines of: we never need cars, horses can carry us and anything we need. We don't need medicine, because there will always be other people to take the place of people who die. Sure, they are true, but there is a "quality of life" argument to be used here. If we can improve the lives of people through the development of machine intelligence, I have no doubt that we will (if only because of market forces). I recall an interview with Kasparov (World Chess Champion) where he said that he didn't know how human beings could go on after being defeated at an intellectual endeavor (chess) by a soul-less machine. The answer should be obvious, we have been beaten by machines for years. They are stronger and more efficient than we are. They haven't been smarter than us (and indeed I don't believe chess is a good indicator of intelligence) but there is no reason to believe that this will not change. >My point is that, even if we reach the point where we can "make" intellig- >ence out of silicon (or gallium arsenide), it will always be more costly >than to have intelligence produced to old way, that is, to make babies. Actually producing a human being is quite difficult, and we know of no engineering shortcuts to speed it up, and no real way of ensuring the quality of the result. As far as commodities, human beings are quite unreliable. >Than why all this fuss about AI? I think this is a just and good effort, >because of the enormous consequences it has already had and it'll have >in the future on the way we understand ourselves. For me, the evolution of machine intelligence is the next step in OUR OWN HUMAN evolution. Language was developed so that knowledge could be quickly and completely transferred from generation to generation, but for machines knowledge could be transferred WITHOUT learning. Once a computer becomes "intelligent" it should be able to utilize learning mechanisms which are manyfold more efficient than our own minds. It is, indeed, stuff to dream about. Mark
akempka@yoda.eecs.wsu.edu (kempka anthony aaron - CS500) (06/13/91)
In article <1991Jun12.194454.19379@pixar.com> markv@pixar.com (Mark VandeWettering) writes: >>It may be a naive idea, but my assertion is that artificial intelligence >>is not worth for any economical results it may produce. In fact, it will > >Similar thinking would be along the lines of: we never need cars, horses >can carry us and anything we need. We don't need medicine, because there I think you're missing his point somewhat. He is speaking of economic results not results of convenience such as cars. I think he has a good point, although a little misinformed. The fact is AI is indeed sought after for its economic benefits. This is much the same as any other software product. If it weren't cheaper to buy an accounting package than to hire an accountant, then the software would not be purchased. A little business economics will show you how expensive a real live person is in comparison to a machine. No vacation, sick leave, insurance benefits, retirement, etc... > >>My point is that, even if we reach the point where we can "make" intellig- >>ence out of silicon (or gallium arsenide), it will always be more costly >>than to have intelligence produced to old way, that is, to make babies. > See above on economics. Any future development of AI is no doubt going to be influenced by economic forces. That is just the way capitalism works. However, this shouldn't be looked upon as a hinderance, but rather a tool that can be used to cultivate some of the current notions of AI. Tony K.