 |
| Exponential curve |
Transhumanists claim that technology is about to reach singularity: exponential growth towards infinity that will allow us to achieve goals such as immortality, strong artificial intelligence, the hybridization of machines and human beings, and much more. They thus make the same mistake made by the members of the club of Rome when in 1975 they predicted that the world's population would grow exponentially and reach catastrophic values by 2020. It is the same mistake made by Malthus at the end of the 18th century, for the growth curves of natural systems are never exponential, but rather follow the logistic curve, as I have pointed out more than once in this blog.
To this
criticism, transhumanists (such as Ray Kurzweil) argue that the human species
is capable of linking two or more logistic curves, so that, even if real
processes follow this curve rather than an exponential, growth would continue
and the singularity could be reached, maybe a little later, but it will come anyway.
Those who say this are dreaming, and show they don’t know mathematics. I show here two linked logistic curves, and compare them to exponential growth. It will be seen that the beginning of the two curves is quite similar, up to the point of abscissa 8.5. However, from that point, the differences are disproportionate. At the point of abscissa 10, the exponential curve reaches the value 3, while the logistics has not yet reached the value 1. At the point of abscissa 30, while the logistic curve reaches the value 2, the exponential would be at about 1446 million. So, claiming that several linked logistic curves equal one exponential, is mathematical folly.
Kurzweil also
says that research in quantum computing should very soon send us into a new
linked logistic curve. This shows that he does not know, or does not want to be
aware, that quantum computing, if achieved, will make it possible to speed up
the resolution of certain types of problems (NP-complete problems), but won’t let us solve new problems, such as I pointed out in another
post. Although it is true that solving this challenge could send us up to a
new logistic curve linked to the previous one, this advance won’t be as
momentous as the media and some scientists would have us believe.
It is evident
from the comparison made that, to approach exponential growth, it would be necessary to concatenate an
enormous number of linked logistic curves, a number tending to infinity, which would require a time tending to
infinity. But is it possible to link a very large number of logistic growths,
or would we find a limit, if we tried?
 |
| Arnold J. Toynbee |
Arnold J.
Toynbee, in his monumental 12-volume Study of History, pointed
out that human civilizations go through a stage of growth, during which they
are subjected to several successive challenges that they usually manage to overcome, although as a rule each challenge solved causes
new problems, and thus opens the way to the next
challenge. Furthermore, he argues
that the number of challenges that a civilization can successfully solve is
usually not large: after defeating three or four, each of which may cost one to
several centuries, it usually fails in the next one and collapses. Toynbee
points out that Western civilization went into political collapse in the early
20th century, with the two world wars. In this he agrees with the
diagnosis of Oswald Spengler, who reached the same conclusion in the
interwar period.
It is true that
the scientific-technological history of our civilization may not exactly
coincide with its political history. However, I don’t think its decline is very
far, as I
have been predicting here. So, announcing that in
the next hundred years we’ll reach singularity (as transhumanists do) is not doing science, but dreaming. At most, these
musings could be the subject of science fiction novels, but nothing more.
Certainly, the
limits of technology are practical, not theoretical, and further advancements
could allow them to be transgressed. But the
hopes of transhumanists in this regard are so exaggerated that they will
probably never be achieved.
Thematic Thread of Science in General: Previous Next
Manuel Alfonseca
No comments:
Post a Comment