|
What if, in the near future, advances in technology allowed us to build massive arrays of supercomputers able to support emergent artificial intelligence. What if such an array was complex enough that we could upload a human mind into it. And what if that mind, freed from organic constraints and with storage, power and speed far beyond what a mere brain could provide, had tools at its disposal -- including programmable, self-assembling nanobot -- to shape the world as it saw fit. Would it be the next stage in human evolution? Or would it be the end of human kind?
Transcendence[^]
It looks like a very interesting movie, but I suspect it will have one of those awful "love conquers all and saves existence yet again" kind of endings.
|
|
|
|
|
Variant of "Wolf In the Fold", Star Trek 1967.
What's old is new again. 
|
|
|
|
|
Not quite: I don't think Depp's character is able to jump from body to body. I think it is more like Arthur C. Clarke's "Dial F For Frankenstein" mixed in with a bunch of transhumanist hooey.
|
|
|
|
|
{dc7c68de-3931-43b9-9102-6dbe3972c5c3} wrote: Variant of "Wolf In the Fold", Star Trek 1967.
I thought you were off by a year, so I looked it up. "Wolf in the Fold" first aired Dec 22, 1967. You were right. +5 from me. I'll go back to my corner now...
Never moon a werewolf.
- Harvey
|
|
|
|
|
Gregory.Gadow wrote: What if such an array was complex enough that we could upload a human mind into it.
Unfortunately, that thought is still the realm of science fiction.
/ravi
|
|
|
|
|
Being a long time fan of science fiction, a reader of cyberpunk and an aspiring SF/F author myself, allow me to fix your post
Unfortunately, that thought is still the realm of science fiction. 
For now, at least.
|
|
|
|
|
Agreed. 
/ravi
|
|
|
|
|
Some of those Youtube comments though...
I'm probably deranged enough to just kill everyone if that happened to me, just because they were bothersome.
|
|
|
|
|
Gregory.Gadow wrote: For now, at least.
For a very long time at least.
AI research has been going on since shortly after computers were invented and the result of all of that has been less than amazing. Other fields have make significant gains with the advent of computers but AI hasn't had any amazing breakthroughs.
|
|
|
|
|
I'm curious though: when you say "amazing breakthroughs", what do you have in mind? If you mean human-like computer minds, then yes, we are still some ways off (though we might be closer than you think [^]); but that's a little like saying there's been no breakthroughs in space exploration since the Sputnik because we don't yet have colonies outside the Earth. That is the goal and we're working on it: but the target problem is so mind-bogglingly hard, the progress we have made so far is remarkable even if we still don't look too close to solving it.
And in some ways the AI challenge is even harder than space exploration. For one, the vision ("people living in space, the Moon and other planets") and requirements (propulsion, life support, radiation shielding etc.) of space exploration are easy enough to agree on; in contrast, see if you can get any three people to agree on what "intelligence" is. Most believe they know it when they see it – they don't [^], and even if they did, this would hardly be enough basis to steer research. With a thousand definitions to pursue, and a thousand strategies to implement each one, it's difficult to even know where "there" is, and much more so to get there.
And yet the AI field has produced many successes over these 60-odd years. LISP has revolutionized the way we look at programming languages. Optimizing compilers produce assembly code of performance comparable (often better) than the output of human experts. Scheduling systems effectively run many industrial plants – the humans are nominally in charge, but in fact all they do is take the orders and nod. Automated VLSI design tools have progressed to a point nobody fully understands the architecture of modern microchips anymore. Automated translation, while still a bit trite, has become good enough for everyday use (I use Google Translate to read messages in Japanese everyday: I know the language well enough to tell if a translation is accurate, but my reading is still too slow, and being able to get an "Engrish" translation at the click of a button is priceless).
Of course, it's often the case that when something works and starts seeing widespread use, it loses the "AI" moniker; sometimes it even becomes its own field. I guess that has a lot to do with the feeling that "AI" is not making much progress – when in fact it's been thriving all along.
"Whereas smaller computer languages have features designed into them, C++ is unusual in having a whole swathe of functionality discovered, like a tract of 19th century Africa."
-- Verity Stob
http://www.theregister.co.uk/2006/05/05/cplusplus_cli/
|
|
|
|
|
Less than amazing?
Seriously? How old are you?
Do you have any idea how many problems were once consider the proper domain of AI that are now in every day use and taken for granted by the masses?
Optical Character recognition? Speech recognition? Facial recognition? Image classification? Autonomous walking robots?
Just because you now have an API that makes these things available to you, doesn't mean you should regard them as mundane.
I suggest you go try to write those functions yourself (without reference to the vast body of AI research that made them possible) and then tell me that AI hasn't produced anything amazing yet.
|
|
|
|
|
And let us not forget Google's smart car, coming to an auto showroom near you (someday).
|
|
|
|
|
|
Gregory.Gadow wrote: we could upload a human mind into it.
That's the one piece of the puzzle that is still nowhere near to being solved.
Veni, vidi, abiit domum
|
|
|
|
|
Oh I dunno - you could download most QA questioner minds into a Z80...and most spammers into an Intel 4004[^]
Those who fail to learn history are doomed to repeat it. --- George Santayana (December 16, 1863 – September 26, 1952)
Those who fail to clear history are doomed to explain it. --- OriginalGriff (February 24, 1959 – ∞)
|
|
|
|
|
I dare say that most spammers could be downloaded to a floppy diskette. More specifically, a single sided, single density 5 1/4.
|
|
|
|
|
I once worked on a system that had 8" floppies. No idea what capacity they were, probably about 256 bytes.
Veni, vidi, abiit domum
|
|
|
|
|
|
Which really was a huge amount in those days.
Veni, vidi, abiit domum
|
|
|
|
|
Side note:
The interior of the Borg cube in STTNG:Q Who[^] has arrays of 8" floppy drives on the walls for the technical look!
|
|
|
|
|
The earliest ones I had were 'hard sectored' 256k. Later, when working with IBM hardware, they used soft sectored 8" floppies which could hold 1 MB.
|
|
|
|
|
You are probably right: 360K should hold a quite a few spammers! 
Those who fail to learn history are doomed to repeat it. --- George Santayana (December 16, 1863 – September 26, 1952)
Those who fail to clear history are doomed to explain it. --- OriginalGriff (February 24, 1959 – ∞)
|
|
|
|
|
I got an ad when I first went to watch the trailer for Transcendence on YouTube. The ad? The trailer for Transcendence. *boom* mind blown.
|
|
|
|
|
Gregory.Gadow wrote: Would it be the next stage in human evolution? Or would it be the end of human kind?
One might as well wonder about magic since current advances would suggest the same likelihood as the scenario you presented.
Beyond that though...
In your scenario is there only one human mind or many?
If many then of course, since more computers is, presumably, more power, then there would be resource wars. If only one or just a few then it would be humans versus computers again in a battle for resources.
One might also presume that the upload brain is faster. Thus even supposing that it was able to find more interesting things to think about, what happens if it becomes bored?
|
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
