INFORMATION TECHNOLOGY SYMPOSIUM
209
Edward Rothstein:
I think the argument Ray makes in the first part of
his book is more startling than the arguments about the evolution of the
computer and "spiritual machines" he makes later on. Ray's argument
is that evolution and the speed of technological change are independent
of particular technologies . This is extremely suggestive because, in his
view, the rate of change has nothing to do with a particular material or
a particular invention, but with the relationship between order and
chaos.
If
indeed this is something that can be formalized, then it changes
the way one has to think about what is happening with technology.
While I'm not convinced that machines will take the path Ray has out–
lined, the first part of his book has convinced me that I have to re-think
the nature of technological change and where it might be leading.
Where I am skeptical, though, is that even if we grant the ever–
increasing speed of technological innovation, Ray, you may be under–
estimating the complexity of the brain itself, which may be many orders
of magnitude more intricate than you think; the neurobiologist Gerald
Edelman calls it the most complex object in the universe. Will it ulti–
mately be reproduced by a machine? I cannot say. But it may not be as
soon as you think.
Ray Kurzweil:
There are different estimates. I use one of the more con–
servatively high ones but let's say I'm off; it doesn't really change things
that significantly.
If
I'm off by a factor of a thousand, that's nine years;
if I'm off by a factor of a million, that's seventeen years; if I'm off by a
factor of a billion, that's twenty-six years. And I don't think I am off,
because there have been a lot of experiments even since the book came
out that suggest the estimates are reasonable. There have been many
experiments to replicate neurons and neuroclusters, not in the form of
mathematically simplified neural nets, but realistically recreated bio–
logical neurons that at least emulate their information; and the salient
processes that affect their functionality, other than digestion and so on,
would show that our models of their functioning are reasonably accu–
rate. But even if I'm off, the power of exponential growth is so great that
even a factor of a billion only means a quarter of a century.
Jacob Weisberg :
Let's take some questions from the floor.
Audience Member:
Ray, it seems that what you are saying is that you're
not in the business of telling us how we're actually going to live in the
future, you're simply trying to predict, with the highest degree of accu–
racy possible, what will be available and what we need to think about.
