Paul Graham has a new essay up, this time making suggestions to CS undergrads about what they should do in college to become a good hacker. In it he makes the following comments:
Professor Conway that I was interested in AI (a hot topic then), he told me I should major in math. I'm still not sure whether he thought AI required math, or whether he thought AI was nonsense and that majoring in something rigorous would cure me of such stupid ambitions.
…
Medieval alchemists were working on a hard problem, but their approach was so bogus that there was little to learn from studying it, except possibly about people's ability to delude themselves. Unfortunately the sort of AI I was trying to learn in college had the same flaw: a very hard problem, blithely approached with hopelessly inadequate techniques. Bold? Closer to fraudulent.
It's still going on. This weeks New Scientist has an article on a chemical medium for computation. Some interesting science, some interesting engineering, and some interesting possibilities. But why, oh why, do they have to claim that it is likely to lead to an artificial brain, complete with self awareness and emotions? On what basis?
This whole AI issue came up last week at a debate on the nature of being human here in sunny Newcastle. Inge Rebergen, whose general approach (that humans are way too arrogant) I had a lot of sympathy for, declared herself to be keen on the idea of strong AI, apparently from the points of view of possibility ,em>and value.
One of my problems with the idea of AI is the assumption that intelligence has any meaning at all separated from the physical experience of being human. Our intelligence is inextricably linked with the physical, chemical, electrical complex of our bodies, including our brain and the nervous system which binds the two and belongs to both. What does it mean to have an "artificial intelligence"? Why do we assume it would be useful to us, that we could communicate with it, having little basis in shared experience? In other words, I'm down on the idea that AI has a well defined meaning, let alone being possible.
My father meanwhile (my parents were up over the weekend for my birthday), who obviously contributed strongly to the views I hold on the matter, voiced the objection that so much money has been and continues to be wasted, steered by whatever means (fraud? ignorance?) into a dream, with a low probability of realisation and little value even if it is realised. Meanwhile, really valuable work on augmenting human intellect — of providing levers to the mind — languishes underdeveloped and underfunded.
Worse, people look at you blankly if you talk about such things: the majority of computer users and even software developers don't understand the importance of the computer as a tool for providing access to powerful abstractions. Where are the tools for building tools?
Comments