 |
| |
 |
| |
|
| |
 |
A few hundred years ago, science was irrelevant to most
peoples’ daily lives. It was mainly a hobby of aristocrats,
in the same vague category as, say, metaphysical philosophy
or jewelry collecting.
Today, science and technology are everywhere. Most people
don’t understand much of it, but they know it’s
important. They use computers and printers and monitors;
they watch TV; they drive cars with fuel-injection engines
and onboard diagnostic computers and fly on airplanes with
autopilots; they take medications with names reflecting
complex molecular structures; they shop at Wal-Mart, which
exists almost entirely to sell various forms of plastic
that that didn’t even exist 50 years ago.
But what most people don’t yet comprehend is that
this is a transitional period. The importance of science
and technology to ordinary life is not going to remain constant
at its present level. Rather, it is going to increase –
and it’s going to increase fast, exponentially or
more so, just as it has been. Computers, plastics, medications
and airplanes are just the beginning.
Technology is going to transform us, redefining what it
means to be human, and not just culturally and psychologically
but physically. Genetic engineering will happen, the protests
of the religious-conservative camp notwithstanding: enhanced
human beings will walk among us. AI will happen, and not
in the 4000 years projected in Spielberg’s recent
movie by that name, but within this century, probably early
on. Computers will be smarter than us in ways more important
than playing chess, factoring large numbers, or landing
jet aircraft. Intelligent and semi-intelligent computers
will help us solve puzzles that have so far eluded us: for
instance, how genomes make cells, how particles make atoms,
and how brains make minds. We will communicate with computers
directly using our brains – “look ma, no hands!”
We will communicate with each other similarly. We will move
back and forth fluidly between the ordinary physical world,
with its trees and roads and love affairs and childcare
centers, and virtual worlds in which anything can happen,
we can wrestle with a molecule or make love to a cloudburst
or solve equations by hybridizing our intuition with that
of an AI mathematician, and it all feels realer than real.
Human nature in some form may persist, but not in the manner
that we are accustomed to it. Science and technology are
leading us into a transhuman age.
This all sounds like crazy science-fiction, and indeed,
history shows that predicting the details of future science
and technology is a pursuit fraught with peril. There will
be limitations I cannot right now foresee, and also revolutions
my paltry human-nervous-system-based creativity is inadequate
to imagine. But even so, I think the qualitative character
of the coming revolution is clear. Vernor Vinge, writing
in the 1970’s, was right to call it a Singularity:
as science and technology advance faster and faster, there
will come a point where, suddenly, they are no longer just
a big part of our world, they are inextricable from our
world. New advances will come faster than we can understand
them. AI’s will create new AI’s that rapidly
outpace their creators.
I write about these topics, not as an outsider, but as someone
who has been right smack in the middle of the ongoing revolution.
After spending about a decade as a scientific theorist,
about 5 years ago I decided to devote my life more fully
to one particular futuristic goal: the construction of a
truly thinking machine, a software program displaying human-level
general intelligence. In 1997 I founded a company, Webmind
Inc., with the twin goals of creating a “Real AI,”
and constructing profitable software products from preliminary
partial versions of this holy-grail end-goal. We made some
interesting progress toward the grand Real AI goal, and
created some pretty impressive software products along the
way, but we never managed to make the firm profitable, and
when the dot-com crash hit, a major funder pulled out at
the last minute, and the company couldn’t recover.
In March 2001, Webmind Inc. dissolved.
Following this personal and financial disaster, a group
of friends from the Webmind Inc. R&D division and I
decided to stick together as a team, and initiate a new
chapter of our quest to build a Real AI (to a great extent,
it had now become our quest, not just mine). We saw the
mistakes we’d made at Webmind Inc., both technically
in terms of the construction of our AI system, and businesswise
in terms of conceiving, creating and marketing products
based on preliminary partial versions of the end-goal Real
AI. We resolved to do it right this time. We are building
a new AI system, the Novamente AI Engine, and we’re
applying it to various practical domains, mainly focusing
on bioinformatics.
But as important as I hope my work will be, I realize that
it’s just one piece of a huge, amazing, unfolding
puzzle. No one scientist will accomplish the Singularity,
no single 21’st century da Vinci or Einstein or even
Goertzel. No one branch of science will accomplish it either.
What is happening is synergetic, cross-disciplinary, international
and emergent.
My aim in these pages is not to convince you that this sort
of revolution is coming. Ray Kurzweil has attempted such
an argument in his recent book The Singularity is Near,
using countless graphs of technological and scientific progress,
showing the hyperexponential growth curves observed in various
industries and dissecting the causes underlying this growth
pattern. I admire his work but don’t aim to replicate
it, or extend it in any direct way. My goal here is more
qualitative, though closely related. It is simply to describe
for you various aspects of contemporary science, which seem
to be pushing in a Singularity-ward direction.
I have focused on those areas of intensely-future-relevant
science that I understand the best, which means mostly computer
science, with a goodly helping of biology, and occasional
dips into other areas such as physics and engineering and
chemistry. But though it’s governed by my own interests,
the coverage of topics is hardly arbitrary, because the
topics I know best are generally things I’ve learned
about precisely because of their central relevance to the
ongoing sci-tech revolution. Some parts of what I’ll
talk about pertain to my own AI research, but the majority
regards to the work of other scientists whom I know, either
personally or through their writings. Of course, in one
brief book I can’t possibly tell you all of what’s
happening in university and industry research labs, garages
and livingrooms all around you. The vast accelerating advance
of science and technology and creative intelligence defies
concise description. But I wouldn’t have bothered
to take the time to write this if I didn’t think I
could do a moderately decent job of approximating the essence
of it all.
Most books these days are written purely for the reader’s
entertainment -- but this one isn’t. I wrote this
book because I thought the ideas contained in it were important
for you to know. And so I’m going to ask something
of you: Read all this, and then think about it. Then if
you want more information, read other books on related topics
– talk to scientists and technologists and ask them
questions. And don’t accept everything I’ve
said: Draw your own conclusions about where things are headed.
The more people understand what is really happening with
science and technology, and what the potential future outcomes
are, the larger will be the segment of humanity that actively
plays a role in the ongoing revolution, rather than passively
being manipulated by it. And in spite of the phenomenal
stupidity that sometimes is associated with large mobs of
humans, on the whole I think that a broader awareness and
conscious participation in our Singularity-ward trajectory
would be a good thing.
One
of the funny things about being a scientific researcher
is that, however high your IQ may be, you wind up feeling
incredibly stupid on a regular basis. You may find yourself
thinking about the same thing over and over again, for months
or years on end, without making any progress at all. And
then when you get the answer, it may seem completely obvious,
and you can’t understand how you could have been so
moronic as not to see it earlier. And then you go out into
the world of ordinary nonscientific people and you realize:
“Wait a minute, I’m actually fairly intelligent
compared to a lot of other people! – I’m not
such an idiot after all! – this research is just really,
really hard for the measly three-pound human brain.”
Given how hard science is even for scientists, explaining
it to nonscientists in a reasonably honest way often seems
an insurmountable challenge. Even a fully technical scientific
paper can never tell the whole story of the research it
purports to summarize. What is fascinating, though, is that
at its best, a nontechnical summary of scientific work achieves
a different kind of meaning from technical scientific papers.
Scientific work focuses on the details – it drills
down so far you’d think it would be impossible to
drill down any further … and then it drills down yet
further. But sometimes it’s valuable not to drill
down, but rather to look back up, and see how each piece
of work relates to everything else – to other bits
of science, and to ideas and events and trends in the nonscientific
world. This is the kind of meaning that nontechnical discussions
of scientific ideas can add. The lack of total scientific
precision allows broader cross-connections and intuitions
to become perceivable. In this sense, I believe, scientific
writing for the nonscientist audience doesn’t have
to be a mere “dumbing-down” of scientific ideas,
it can be a “contextualization” of scientific
ideas, a deepening of scientific ideas in a different direction
from the direction they’re taken by scientific research
work.
(Okay, okay, so that’s a rather lofty perspective.
If I’m lucky I’ve achieved this high-falutin’
ideal 30-40% of the time in the following pages. But at
least you know what I’m aiming for!)
Of course, contextualizing science is one thing in the context
of a narrow scientific topic such as laser physics or genetics;
it’s quite another in the context of a synergetic
mix of scientific topics whose common theme is their potential
to transform the nature of humanity. I have bitten off a
rather substantial task here. But anyone who knows me could
tell you that unambitiousness is one of the few things not
included on my long list of flaws. After all, the primary
goal of my scientific career has been the creation of a
computer program with human-level general intelligence –
and after 15 years I’m still at it, and still optimistic
(and I’ll tell part of that story in Chapter 3).
One aspect of popular science books that I have mixed feelings
about is the incorporation of biographical information.
Knowing the social context of an idea’s conception
and development can be both fascinating and clarifying;
and the adventure of scientific discovery can be gripping
and exciting and well worth dramatic recounting. But on
the other hand, I’ve read many books that told me
far too much about the story of a scientific discovery,
and not nearly enough about the discovery itself. I’ve
tried to walk the fine line here, in this regard. The ideas
described in these pages involve countless human stories,
and I’ve told a small selection of these, introducing
biographical information when it seemed particularly useful
for bringing out some aspect of the sci-tech points in question.
In these passages, rather than focusing on the already famous,
I’ve made an effort to illustrate the breadth of scientific
insight that is going into the sci-tech revolution, by recounting
bits and pieces of the lives and intellectual adventures
of some lesser-known individuals who have nonetheless made
outstanding contributions.
My own story – the rise and fall of my company Webmind
Inc., the gathering-together of an international team of
mad scientists, programmers and financiers to focus on building
a thinking machine – is not without its dramatic interest
either, but I have only touched on it briefly here and there
in these pages, when it seemed particularly apropos. If
I ever have the time I’ll write a whole book on the
Webmind Inc. story – a fascinating tragicomedy, to
be sure. But this is not that book. Here the ideas are the
focus. I think they’re dramatic enough.
One big part of the overall task of contextualizing science,
is placing scientific results in a social and ethical context.
Obviously, this is a particularly critical task when one
considers scientific work of the sort I’m discussing
here -- work that promises to, in time, completely redefine
what it means to be human (and indeed, whether the “human”
category continues to exist at all).
My perspective here is rather distant from the current ethical
debates on biotechnology, which center on such things as
the ethics of human cloning or genetically engineered produce.
From my own deep-futurist point of view, the correct outcome
of these debates is painfully obvious. If genetically engineered
produce can help feed the hungry people of the Third World
– and it can -- we should pursue it unreservedly.
On the other hand, the fuss about cloning seems
to me to be almost entirely due to peculiar religious notions.
I have not seen anything resembling a convincing argument
that human cloning will lead to great dangers. While I have
great respect for the deep spiritual experience that lies
at the heart of all religions, it seems to me that religious
superstitions have been and will be the cause of a lot more
pain and suffering than could ever come from human cloning.
I think the ethical debates over “frankenfood”
(I love that expression – for a while I’ve been
talking about starting a rock band composed entirely of
biologists and called “Frankenfood Buffet”),
human cloning, stem cells and related topics will die down
within 10 years or so, as these relatively simple kinds
of biotechnology become commonplace. The really profound
and important ethical debates will be the ones that come
after – the ones to do with advanced genetic engineering,
man-machine synthesis, and the nature of posthumanity.
There is a real risk that one day genetic engineering will
lead to the creation of a genetically-superendowed wealthy-nation
elite, with the vast majority of the world unable to afford
genetically enhanced children, just as they now cannot afford
computers or Gameboys or GPS widgets or cutting-edge pharmaceuticals.
There is a real risk that the wealthy will upload into a
digital world, leaving the masses in a polluted, increasingly
unlivable physical-world ghetto. There is also a risk that,
somehow, in creating these exciting technological modifications
to ourselves, we will lose the passion, the feeling, the
essence of being human – that we will fashion for
ourselves a more perfect but less intense and genuinely
fulfilling reality. Perhaps the pharmacological elimination
of suffering and the creation of viscerally satisfying virtual
worlds will make life too easy, destroying the rich if sometimes
difficult texture of everyday being that we take for granted.
It is my view that these ethical nightmares can be averted,
but it’s not entirely clear that humanity as a whole
will have the will to avert them.
Some of those who wish to stop human cloning and genetically-modified
food may feel as they do because they view these things
as steps along the way toward more dramatic and dangerous
things. I have some respect for this perspective. But I
believe that the quest to stop science and technology developing
– Bill Joy’s “relinquishment” –
is doomed to fail. A pessimistic view would say: “Pandora’s
box cannot be closed again.” But it’s better,
in my view, to say: “It’s clear the future will
involve these technologies, so let’s direct our energies
toward maximizing the chance that these technologies will
be used in a positive way.” It is entirely possible
to use advanced computing and biotechnology to make human
experience richer, deeper and better, for all humans. The
more people realize this goal is important, the more likely
it is to come about.
Genetically modifying humans so that they can more effectively
hybridize themselves with computers, more effectively jack
into virtual realities, share thoughts and feelings with
each other directly rather than through the narrow-bandwidth
medium of language. It sounds science-fictional, but yet,
it seems to me extremely probable that this is 21’st
century science, not 22’nd century. The research of
today points in this direction just as clearly as, say,
the common existence of radio and cinema once pointed towards
television. Significant new inventions will be required,
but it is abundantly clear that the human race possesses
the collective brilliance to make them.
In these pages I’ll tell you about what scientists
are doing today at the boundary of biology and computing,
and what I believe they’ll be doing tomorrow, and
how it fits into a bigger picture of “posthumanist
philosophy.” It was over 100 years ago that Nietzsche,
in his Zarathustra, said “Man is something that must
be overcome.” He did not envision that silicon chips,
gene chips, PCR and so forth would be the mechanisms of
this self-overcoming – but his insight was dead-on
nonetheless. We are well along in the process of overcoming
ourselves, not through the visionary trances of prophets
like Zarathustra, but through the workaday R&D activity
of thousands of scientists worldwide, studying gene expression
data analysis, neurocomputing, advanced computer architecture,
anti-aging pharmacology, and a dozen other cutting-edge
disciplines. It is humanity as a whole, not any particular
individual, that is in the visionary trance, courtesy of
the socially-psychoactive substance called science, and
heading straight toward a major transformation of the soul.
It’s quite a drama that’s unfolding, and we’re
right smack in the middle of it – and within some
fairly broad parameters, we get to choose the roles we play.
|
|
|
