An AI going to be just as clueless as we are, regardless of how quickly it thinks.
What do you mean by this? "Thinking faster" isn't the only way to make an AI smarter.
Perhaps it could truly turn into some kind of exponentially growing demigod if we gave it an autonomous body of some sort, like giving self-replicating machines a collective consciousness.
Or just gave it access to the Internet. Lots of people out there who can be cheated to give you stuff: it doesn't even require superhuman intelligence, as any successful con man could tell you. And cheating is just one example of how it could achieve that.
But that notion is too silly to even consider: any healthy intellect would just kill us all - as evident from nature; and one burdened by some kind of Asimov's Laws is bound to be a sad, dependent cretin.Nature isn't a reliable guide: an AI would be engineered by humans, not by evolution. Even if you did
( ... )
My reply slightly exceeds the character limit, so it's in two parts - sorry.
"Thinking faster" isn't the only way to make an AI smarter.How else, then? I see two variables that go into determining intelligence: the volume of data provided, and the speed at which this data can be parsed to reach conclusions from it. It can be said that human civilization is a single intelligent entity for which both these variables have been growing since the very start. Yet, our civilization now seems no smarter than it was during the Roman Empire. We think many new thoughts, and play with many new toys, true. But the basic facets of existence have not changed. The lives of ants, the lives of rats, and the lives of men are all driven by the same sets of problems - entropy, pain, curiosity. Do you honestly think a superior intellect is going to make more sense of this? I haven't a single doubt that it'll think enormous thoughts that we can't even imagine, and discover how to make toys that far surpass our current knowledge of physics. But is it
( ... )
I think gene therapy is more powerful than you think. ... Granted, it will not heal the pathology that has already accumulated, but with luck, it will prevent any further damage from occuring, as any further waste material will be eliminated before it can damage reach levels where it causes more damage. ... With a bit more luck, the damage that has already been done will then begin to heal as the cells replace themselves normally.Oh, I haven't the slightest doubt that gene therapy is powerful - except that it's powerful on a cellular level, not an organ level. I'm going to try illustrating this with a metaphor. Suppose there is a park that's littered with leaves and trash. Plastic bags and so forth. We want the park to be a clean, sunny lawn; we use this lawn for picnics. It's impossible to rake that debris, because we don't have rakes; we only have huge bulldozers (the size of Godzilla). By planting a different type of tree growing in that park (say, a pine) it'll get better, since pines don't shed as much as deciduous
( ... )
I'd just like to remind you, an organ is made from individual cells that keep on dividing throughout our lives. If you introduce a mutation to one of the cells, it's carried into multiple descendants - the same way as a tumor keeps on growing. Since you have new cells that are being born, and the size of the organs doesn't vary that much, this means that the old ones die at roughly the same speed. So, if you insert an identical mutation to multiple cells in an organ, let's say by injecting a large amount of engineered viruses, at some point those new cells have taken over the whole organ. If these cells are any good, you can consider that the organ has been replaced, bit by bit, by cells that are engineered to do their function and something extra
( ... )
Unlike you, I think the point of singularity is in unpredictability of events after the point of singularity by conventional means of leveraging the advance of a civilization. The world after singularity might as well be hell or heaven. The point I believe, is in unpredictability from rapid increase in capacity of the human civilization, and Ray Kurzweil is suggesting that the engine of drive beyond such rapid expansion of human capacity would have to be artificial intelligence of some kind... So at least in mechanic and spirit, the idea of singularity would be different from the idea of a rapture in that positive outcome is not somethinng guaranteed, although I do agree that there are just too many similarities for some of us to feel comfortable with the notion
( ... )
Well, this is how I view "Technological Singularity": a lifeform that is substantially more intelligent than us (strong AI) evolves very quickly
( ... )
Good points. The number of life-like systems with built in death mechanism is outnumbered by the ones that doesn't have any such features. As for circulaatory degradation being the primary cause of natural death, I'm afraid I don't have any expertise on the matter. I'd be very interested in learning further. Can you show me your proof?
As for circulaatory degradation being the primary cause of natural death, I'm afraid I don't have any expertise on the matter. I'd be very interested in learning further. Can you show me your proof?
Well yes, I've linked this a few replies up; it's a list of the leading causes of death on Wikipedia. Here it is again. I'm not the least bit surprised; almost every dead relative I can think of has died from cardiovascular disease of some variety.
Comments 31
What do you mean by this? "Thinking faster" isn't the only way to make an AI smarter.
Perhaps it could truly turn into some kind of exponentially growing demigod if we gave it an autonomous body of some sort, like giving self-replicating machines a collective consciousness.
Or just gave it access to the Internet. Lots of people out there who can be cheated to give you stuff: it doesn't even require superhuman intelligence, as any successful con man could tell you. And cheating is just one example of how it could achieve that.
(My articles "14 Singularity objections answered" and "Why care about artificial intelligence?" might of some use to you.)
But that notion is too silly to even consider: any healthy intellect would just kill us all - as evident from nature; and one burdened by some kind of Asimov's Laws is bound to be a sad, dependent cretin.Nature isn't a reliable guide: an AI would be engineered by humans, not by evolution. Even if you did ( ... )
Reply
"Thinking faster" isn't the only way to make an AI smarter.How else, then? I see two variables that go into determining intelligence: the volume of data provided, and the speed at which this data can be parsed to reach conclusions from it. It can be said that human civilization is a single intelligent entity for which both these variables have been growing since the very start. Yet, our civilization now seems no smarter than it was during the Roman Empire. We think many new thoughts, and play with many new toys, true. But the basic facets of existence have not changed. The lives of ants, the lives of rats, and the lives of men are all driven by the same sets of problems - entropy, pain, curiosity. Do you honestly think a superior intellect is going to make more sense of this? I haven't a single doubt that it'll think enormous thoughts that we can't even imagine, and discover how to make toys that far surpass our current knowledge of physics. But is it ( ... )
Reply
I think gene therapy is more powerful than you think. ... Granted, it will not heal the pathology that has already accumulated, but with luck, it will prevent any further damage from occuring, as any further waste material will be eliminated before it can damage reach levels where it causes more damage. ... With a bit more luck, the damage that has already been done will then begin to heal as the cells replace themselves normally.Oh, I haven't the slightest doubt that gene therapy is powerful - except that it's powerful on a cellular level, not an organ level. I'm going to try illustrating this with a metaphor. Suppose there is a park that's littered with leaves and trash. Plastic bags and so forth. We want the park to be a clean, sunny lawn; we use this lawn for picnics. It's impossible to rake that debris, because we don't have rakes; we only have huge bulldozers (the size of Godzilla). By planting a different type of tree growing in that park (say, a pine) it'll get better, since pines don't shed as much as deciduous ( ... )
Reply
Reply
Reply
Reply
Reply
Well yes, I've linked this a few replies up; it's a list of the leading causes of death on Wikipedia. Here it is again. I'm not the least bit surprised; almost every dead relative I can think of has died from cardiovascular disease of some variety.
Reply
Leave a comment