| Both sides previous revisionPrevious revision | |
| public:t720-atai-2012:what_is_agi [2025/04/27 17:52] – thorisson | public:t720-atai-2012:what_is_agi [2025/04/27 17:53] (current) – [What is General Machine Intelligence?] thorisson |
|---|
| By now it should be clear that the "g" in "GMI" (general machine intelligence) is an attempt to put back the emphasis on holistic intelligence, in the pursuit of artificially intelligent systems. It is there to re-invigorate the hopes and dreams of the founding fathers of A.I., such as Alan Turing, Marvin Minsky, Alen Newell, John McCarthy, and others, who thought that it might be possible to challenge human intelligence with a man-made information processing machine. Sure, they got some or most of their methodologies, assumptions, and predictions wrong, but that is inevitable in the early days of any scientific field. And we still agree with their main vision – that this goal is possible to achieve. However, we must choose our methodology carefully, hone our tools thoughtfully, and most importantly: We must not be tempted to simplify the thing we are studying -- //intelligence// -- so much so that it starts to differ significantly from the very phenomenon that got us interested this pursuit in the first place -- or even worse, starts to look like //something else entirely//. | By now it should be clear that the "g" in "GMI" (general machine intelligence) is an attempt to put back the emphasis on holistic intelligence, in the pursuit of artificially intelligent systems. It is there to re-invigorate the hopes and dreams of the founding fathers of A.I., such as Alan Turing, Marvin Minsky, Alen Newell, John McCarthy, and others, who thought that it might be possible to challenge human intelligence with a man-made information processing machine. Sure, they got some or most of their methodologies, assumptions, and predictions wrong, but that is inevitable in the early days of any scientific field. And we still agree with their main vision – that this goal is possible to achieve. However, we must choose our methodology carefully, hone our tools thoughtfully, and most importantly: We must not be tempted to simplify the thing we are studying -- //intelligence// -- so much so that it starts to differ significantly from the very phenomenon that got us interested this pursuit in the first place -- or even worse, starts to look like //something else entirely//. |
| |
| P.S. Research in AI sits on two pillars, engineering and science. In engineering the goal is to follow a //model// -- to make the world behave according to the blueprint, be it a bridge, a house, a computer, a network, or something else. Science strives to //discover// the model, which is //not known//. These approaches work together in AI, but if your main goal is not science -- the uncovering of new knowledge -- then it is perfectly acceptable to build a system with a practical purpose. If you're a scientist, it is perfectly acceptable to use any and all engineering tools and tricks in your search for knowledge. Just don't confuse the two end goals, it will confuse everything and everyone, and while you may be able to get away with it (publish a lot of papers, get awards, get rich even), you may confuse the field you belong to -- the youngsters coming into the field seeing your legacy -- because you make them think that an unclear focus is the norm. This is, I'm afraid, very much the current state of affairs in the field of AI. | Research in AI sits on two pillars, engineering and science. In engineering the goal is to follow a //model// -- to make the world behave according to the blueprint, be it a bridge, a house, a computer, a network, or something else. Science strives to //discover// the model, which is //not known//. These approaches work together in AI, but if your main goal is not science -- the uncovering of new knowledge -- then it is perfectly acceptable to build a system with a practical purpose. If you're a scientist, it is perfectly acceptable to use any and all engineering tools and tricks in your search for knowledge. Just don't confuse the two end goals, it will confuse everything and everyone, and while you may be able to get away with it (publish a lot of papers, get awards, get rich even), you may confuse the field you belong to -- the youngsters coming into the field seeing your legacy -- because you make them think that an unclear focus is the norm. This is, I'm afraid, very much the current state of affairs in the field of AI. |
| |
| \\ | \\ |
| \\ | \\ |
| //2020(c)K.R.Thórisson// | //2025(c)K.R.Thórisson// |
