Will artificial intelligence mean the end of humanity
someday? That’s the rather
speculative argument in Matt Miller's Washington Post column last month. Miller suggests that indeed many of our
brightest technological gurus are arguing that computer intelligence is on the
cusp of surpassing our own. He
takes his main case from Ray Kurzweil, the best known promoter of the
Singularity, when computers are powerful enough to blend in with our own and
forever change our nature. Lately a number of columns have sprouted, that
suggest that the truly eventful, if not terrifying, technological change in our
time will be the shift when computers become truly intelligent. Blogger Kevin Drum echoed
similar ideas similar ideas just the other day.
Kurzweil, the most famous and well-known
technological seer embraces Turing’s argument that while we cannot be sure that
computer’s will one day be sentient, the question will fade in importance once
the power of computers enable them to perform any task we can, including
creating art and novels.
Miller, rather early in his column poses the question: “What happens when we share the planet
with self-aware, self-improving machines that evolve beyond our ability to
control or understand? Are we
creating machines that are destined to destroy us?"
Well, let’s slow down and pause on
the vision of Terminators offing humanity. Asking what happens with machines become self-aware is kind of like asking what happens when we eventually invent a time machine where we can
go back in time and enable white supremacists to give Hitler the bomb. Why aren’t we worried about
that? Well, while it might work as a plot for science fiction, there
is nothing we know from physics that suggests that’s remotely possible. And similarly, there is nothing in
physics that suggests that computers will one day soon be sentient.
I
doubt that these authors (and even Kurzweil) believe that their current desktop
computers are sentient. Yet these
use fundamentally the same technologies as those in Alan Turing’s day. They are much faster and hold much more
data, to be sure. And of course
there is much more interconnectivity between computers today, ranging from your desktop
to your cellphone to the webservers that create the internet. When you ask Siri on your iphone for
directions to a nearby coffeeshop, it can even seem like an intelligent entity
is giving you directions. But the
Apple engineers who designed Siri do not argue that Siri is sentient in the
least. Not even a little sentient.
The
only way computers will someday experience true sentience is if a radical new
technology is developed and implemented that is currently nowhere on the
books. That is, no one truly has
any idea how consciousness emerges from matter. This is what philosophers of mind call the Hard Problem of
consciousness. Nothing in the laws
of physics as we know them today suggests how vast, complex collections of
non-conscious particles can somehow become conscious. Computers are becoming more powerful tools. One day soon, we might have
sophisticated robots that seem intelligent. However, they will not be any more truly sentient than Siri.
Computers will continue to advance and become more powerful
tools that chance our world in profound ways. This is an important trend that does justify much careful thought. But computers will not be writing plays that rival Shakespeare because, among many reasons, they do not possess emotion. And that’s because they do not and cannot experience
anything.
UPDATE: Well, robots may not become sentient, but they are advancing. In South Korea, they will soon have their own themepark!! Check it out!
UPDATE: Well, robots may not become sentient, but they are advancing. In South Korea, they will soon have their own themepark!! Check it out!
