| | Here are some interesting [selected] excerpts from the Wiki on The Singularity is Near:
2020s **************** The threat posed by genetically engineered pathogens permanently dissipates by the end of this decade as medical nanobots¡ªfar more durable, intelligent and capable than any microorganism¡ªbecome sufficiently advanced.
A computer will pass the Turing test by the last year of the decade (2029), meaning that it is a Strong AI and can think like a human (though the first A.I. is likely to be the equivalent of a kindergartner). This first A.I. is built around a computer simulation of a human brain, which was made possible by previous, nanotech-guided brainscanning. **************** 2030s **************** Mind uploading becomes possible. Using brain nanobots, recorded or real-time brain transmissions of a person's daily life known as "experience beamers" will be available for other people to remotely experience. This is very similar to how the characters in Being John Malkovich were able to enter the mind of Malkovich and see the world through his eyes.
The same nanotechnology should also allow people to alter the neural connections within their brains, changing the underlying basis for the person's intelligence, memories and personality. **************** 2040s **************** Human body 3.0 (as Kurzweil calls it) comes into existence. It lacks a fixed, corporeal form and can alter its shape and external appearance at will via foglet-like nanotechnology. Organs are also replaced by superior cybernetic implants. ****************
2045 **************** $1000 buys a computer a billion times more powerful than all human brains today. This means that average and even low-end computers are hugely smarter than even highly intelligent, unenhanced humans.
The Singularity occurs as artificial intelligences surpass human beings as the smartest and most capable life forms on the Earth. Technological development is taken over by the machines, who can think, act and communicate so quickly that normal humans cannot even comprehend what is going on; thus the machines, acting in concert with those humans who have evolved into postbiological cyborgs, achieve effective world domination. The machines enter into a "runaway reaction" of self-improvement cycles, with each new generation of A.I.s appearing faster and faster. From this point onwards, technological advancement is explosive, under the control of the machines, and thus cannot be accurately predicted.
With the entire universe made into a giant, highly efficient supercomputer, A.I./human hybrids (so integrated that, in truth it is a new category of "life") would have both supreme intelligence and physical control over the universe. Kurzweil suggests that this will open up all sorts of new possibilities, including manipulation of the physical constants, inter-dimensional travel, and controlling the fate of the universe. ****************
In the 2020s, medical nanobots conquer infectious disease. This seems implausible because it treats microbiological life as static. But microorganisms replicate fast. E. coli -- given nutrition -- doubles every 20 minutes (would cover the whole earth in just 3 days), and each doubling could involve mutation. By treating life as something static, and technology as something advancing, it appears that technology would dominate or conquer life -- but that view is too simple. Also in the 2020s (in 2029, the Turing Test will begin to fail to differentiate between human and computer). Now, I believe that there are computers right now that can existentially pass the Turing Test when the human investigator is a child, or mentally-undeveloped person. Some people will be fooled, but that is because they are, themselves, fools (gullible, credulous). If you plant a 6 year-old kid in front of a computer screen and tell her to start typing to the "person" on the other end -- and then you ask her if the "person" she was communicating with is a human or not -- then she'll be tricked. But the $64,000 question -- of which the child is ignorant -- is that you have to get your interlocuter to communicate what it feels like to be human: 1) machines will always fail at this task 2) even some humans -- mentally undeveloped -- will fail at this task In the 2030s, we're supposed to be able to upload minds, but there isn't even consensus on what a mind is. That's arbitrary and, therefore, wrong. Also in the 2030s, we'll be able to "be" other people (to have their inner experiences as if we were them), but this is simplistic. Folks are complex. There's more to an inner experience than some well-written software. Even all of the software in the known world would not be able to conjure up something like this. Also in the 2030s, we can change ourselves into being "non-ourselves" -- by fiddling with neural connections. I could be James Dean, for instance (or not appreciably different from James Dean). This seems far-fetched. In the 2040s, we become "Transformers" and can live in 100% mechanized, synthetic material. But again, this treats life as being simpler than it is. To repeat, we can't even make a living leaf ex nihilo (without starting with living material from a tree or a leaf). In 2045, when we hit the Singularity, we transcend metaphysics and identity. I don't buy that. Ed
(Edited by Ed Thompson on 8/12, 10:27am)
|
|