IEEE Spectrum September, 2008 - 5

forum
and, consequently, its
input--output behavior.
Once we have achieved
this-assuming it will
ever happen-there
will be no economical
reason to build exact
replicas of human minds.
Instead we would
build intelligent agents
for very specific tasks.
These agents would at
least be alien to human
minds, and there is no
reason to believe that
their intelligence would
not continue to grow.
Philipp Wolfrum
IEEE Student Member
Frankfurt, Germany
Letters do not
represent opinions
of the Ieee. short,
concise letters are
preferred. they
may be edited for
space and clarity.
Additional letters
are available online
in "And More Forum"
at http://www.
spectrum.ieee.org.
Write to Forum,
IEEE Spectrum,
3 Park Ave., 17th
Floor, New York, NY
10016-5997, U.s.A.;
fax, +1 212 419 7570;
e-mail, n.hantman
@ieee.org.

10

More Notes oN
the siNgularity

W

hile Robin
Hanson raises
some thought--provoking
points in his article
"Economics of the
Singularity" [June],
there is one argument I
strongly disagree with.
Hanson assumes that the
first intelligent machines
will be exact copies of
human brains. I think
that most researchers
in neuroscience or
intelligent systems
would agree that this
will not be the case. We
are far from having
the technology that
could scan and exactly
reproduce the functional
structure of a system
as complex as the brain.
But even if we could, a
mere static copy of a
brain would not display
humanlike behavior.
For this we also would
have to reproduce
the self--adaptive
qualities of the brain,
which continuously
modify its structure

NA * iEEE SpEctrum * SEptEmbEr 2008

I

greatly enjoyed
reading about the
singularity but wonder
why its advocates
ignore the fundamental
difference between
computer models and
reality. For example,
Christof Koch and Giulio
Tononi ["Can Machines
Be Conscious?" June]
correctly observe that
consciousness is part
of the natural world,
but their conclusion
that consciousness
functions only through
mathematics and logic is
false. While a computer
model may simulate the
hydraulic pressure in
a heart, this simulated
pressure can't pump real
blood. In the same way, a
computer may simulate
certain aspects of
consciousness, but this
artificial intelligence will
never be conscious itself.
The human brain
operates on reality, not
numbers. Numbers
are only attributes of
reality, not the whole of
it. Indeed, the logician

Kurt Gödel proved
in 1930 in his famous
incompleteness theorem
that mathematics is
incomplete; often you
must go outside of
mathematics to arrive at
truth. Gödel's theorem
is the equivalent of
saying there are valid
computations that will
never be complete. The
only possible response of
a robot to much of reality
is, "It does not compute."
To be human means to
experience life. While
simulating life and its
many processes can
bring wonderful benefits
to mankind, confusing
simulations with reality
is harmful and absurd.
Zoltan Cendes
IEEE Fellow
Pittsburgh

T

he Consciousness
Conundrum" [June]
was a well--thought--
out, well--documented
article on the fantasy of
the singularity. British
neurobiologist Steven
Rose, one of the experts

cited, asserts that our
minds require the
medium of "a social,
crafty, emotional, sex--
obsessed flesh--and--blood
primate," which almost
says it all. I would add
that an essential part of
our being is our mortality.
It provides a large part of
our motivation for many
of our most defining
achievements, including
child rearing, teaching,
writing, art, science, and
dreams of the singularity.
Mortality also clears the
Darwinian deck for the
next generation to evolve.
If the lives of all our
ancestors had been twice
as long, the sun would
have gone into overheat
mode long before
something like us could
evolve and do something
about it. So let's not be
too quick to frustrate the
life--and--death process
that has been such an
important part of us and
our ancestors.
Michael Mallary
IEEE Fellow
Harmony, Pa.

CorreCtioN

we would like to credit Henry
markram, of the Brain mind institute at ecole
polytechnique Fédérale de Lausanne, as the
original source of the images on the contents
page and in "the consciousness conundrum"
[above] in our June issue. markram's work on the
Blue Brain project was the source of the images.
www.spectrum.ieee.org
http://www http://spectrum.ieee.org http://www.ieee.org http://www.spectrum.ieee.org
Table of Contents for the Digital Edition of IEEE Spectrum September, 2008
