Vernor Vinge, science fiction author, computer scientist, and retired mathematics professor, is the inventor of the term 'singularity' as applied to a future point of unprecedented technological progress, caused in part by the ability of machines to improve themselves using artificial intelligence. It's a fascinating and controversial concept, one which could directly relate to molecular manufacturing.
If nanofactory technology makes it possible to embed the equivalent of today's supercomputing power into a device as small as a grain of sand, and to network thousands or millions of those devices into a super-supercomputer, then it appears we'll easily have all the hardware required for a brute force approach to superintelligence.
But is that all that's needed? What about software? What about a better understanding of how intelligence and consciousness emerge from within the human organism, including brain, glands, and sensory organs? These questions still remain unanswered.
In the short interview above, Vinge seems to imply that the singularity, as he envisions it, could be reached quite soon, and likely no later than 20 or 30 years from now. Back in 1993 -- exactly 15 years ago -- he confidently predicted that "Within thirty years, we will have the technological means to create superhuman intelligence. Shortly after, the human era will be ended."
If the human era is indeed near an end, the obvious question to ask is "What comes next?" "What happens to us?"
Does the end of the human era mean the end of humans? If so, is that a good thing or a bad thing? Is there anything we can do now to influence the direction a singularity might follow? Michael Anissimov has some thoughts on these matters at his blog. (I am less confident than he is about the ability of mere humans to guide the future desires of a posthuman superintelligence.)
Perhaps all of this, however, is much ado about nothing. It could be that the future will in fact end up being a lot like the present, mostly just more of the same. Progress will roll on, slowly and bumpily, without any cataclysmic transformations. Human society will change and adapt, gradually, and muddle through as we always have. Even the much-feared (by some, including us) effects of global climate change will prove to be less disastrous and more manageable than most scenarios have predicted.
It's impossible to know in advance, of course, what the future will bring. (That's why it's called the future!) So we suggest hoping for the better while preparing for the worse -- which includes extrapolating from today's trends, modeling different possible outcomes, and examining potential impacts of emerging technologies. That's what we consider the responsible approach, and that's what we try to do here at CRN.
I believe a "cold intelligence" IS possible, as Vernor Vinge does. I also believe that human evolution is occurring under the radar at a rapid pace.
So there will be two evolved intelligences to face in the future. Which one wins? I don't know, and I won't be here to see it, absent some revolutionary advance in nanotech that rebuilds human body infrastructure.
Which one wins? Who cares?
The real question is does the human races deserve the chance to continue to compete?
Does an evolved AI take off on it's own, or do we find a way to work together, humans and "others," for the betterment of both?
Posted by: Mediaman | August 15, 2008 at 11:18 AM