Computers with super human intelligence may lead to a doomsday ...

heartlustΗλεκτρονική - Συσκευές

2 Νοε 2013 (πριν από 4 χρόνια και 6 μέρες)

200 εμφανίσεις

Technological Singularity

Preventing Post Humanity

By Alex Jarstad

Introduction to Technological
Singularity


Technological singularity is the point at which
computers achieve artificial intelligence that
surpasses the intelligence of humans
.


Scientists generally agree that this will occur
before the year 2030
.


The term Singularity was first proposed in
1993 by Theorist
Vernor

Vinge
.

Technological
Singularity



Let an ultraintelligent machine be defined as a machine that can
far surpass all the intellectual activities of any man however clever.
Since the design of machines is one of these intellectual activities,
an ultraintelligent machine could design even better machines;
there would then unquestionably be an ‘intelligence explosion,’
and the intelligence of man would be left far behind.
Thus the first
ultraintelligent machine is the last invention that man need ever
make
.”


I.J. Good, 1965

Historical Background


3000 B.C. The
Chinese invent a counting device known as the “abacus”.



1623 A.D. Wilhelm
Schickard

develops the first automatic calculator.



1679 A.D.
Gottfried Wilhelm Leibniz develops the modern binary system
.


1938 A.D.
As
a result of World War II starting, the United States Army begins
research towards the development of the
ENIAC
.



1947 A.D.
The transistor is invented at AT&T Bell labs.


1965 A.D. Gordon
Moore makes the discovery that the number of transistors that
can be placed on a circuit board and still be cost efficient will double every 24
months.



1988 A.D. Hans
Moravec

creates an early projection for when the intelligence
explosion will occur.

Moore’s Law


The number of transistors that can be placed on an
integrated circuit will increase exponentially
approximately every two years in relationship to cost.


Has held true to the test of time (Twist, 2005).

http://www
-
ece.ri ce.edu/~jdw/fi gs/2n3904b.jpg

http://visual.merriam
-
webster.com/images/science/physics
-
electricity
-
magnetism/electronics/packaged
-
integrated
-
circuit.jpg

http://pop.pcpop.com/upimg2/2004/4/2/282073576.jpg

Moore’s Law

http://singularity.com/images/charts/MooresLaw.jpg

Moravec’s

Projection

http://www.frc.ri.cmu.edu/~hpm/talks/revo.slides/power.aug.curve/power.aug.gif

What is intelligence?


Human intelligence


Animal intelligence


Artificial intelligence

http://l esbi anpi ratequeen.fi l es.wordpress.com/2009/02/intel ligence.jpg

The Human Brain


According to Steven Pinker, author of
How the Mind
Works
, “The mind is a system of organs of
computation, designed by natural selection to solve
the kinds of problems our ancestors faced in their
foraging way of life, in particular, understanding and
outmaneuvering objects, animals, plants, and other
people (1997
).”


Physical Components


Neurons


Specialized Brain cells required for functioning (
Bear et al.,
2007).


Glia

(
Glial

cells)

http://www.utexas.edu/courses/bi o365r/Images/neuron.JPG

Functional Areas of the Brain

http://medi a
-
2.web.bri tanni ca.com/eb
-
medi a/32/99532
-
004
-
2B7BE4E6.jpg

Complexity of human thought


Years of mental development prove to be
adaptive (Wundt, 1897).


Cortical Brain functioning allows for
hypothesis testing
(Bennett & Hacker, 2005).

Computer Intelligence


Can computers think?

The Turing Test

http://www.al anturi ng.net/turi ng_archi ve/graphi cs/turi ngtest.gi f

Implications of Turing Test


An error
in identifying the human exhibits the
computer’s ability to think and trick the participant
.


“The
only way to know that a
man
thinks is to be
that particular man (Turing, 1950
).”


In the years that followed the work of Turing the
primary question of AI turned from “Can computers
think?” to “How well can computers think?” (
Hayles
,
1999).

Computer thinking vs. Human thinking

Computer Thinking


Serial Processing


Can analyze
preprogrammed
problems

Human Thinking


Parallel Processing


Can analyze novel problems



http://si ngul ari ty.com/images/charts/thumb_SuperComputers.jpg

Completely Autonomous AI


Understand Full Spectrum of Human Emotion


Experience Emotion


Be able to recognize images and language


Sense of self preservation

Recent Developments


Quantum Computers


Building The Matrix (
Castelvecchi
, 2008).


Parallel Processing Computers


Cray and Intel multiyear deal
(Thomas, 2008).


Self Aware Artificial intelligence


NaturalMotion

Euphoria software
(
NaturalMotion
, 2008)


Socializing Robots


Robots develop unique personalities based on
interactions with humans
(
Ceurstemont
, 2009).


Facial recognition software


Facial expression recognition


The Singularity is Near


Kurzweil

(2005) on reverse engineering the brain


Because
human intelligence exists means it is possible to replicate it.


Forms of artificial intelligence already exists within search
engines, video games, and cell phones.


One
of the biggest setbacks towards developing AI is that
computers right now cannot process language the way
humans can. However, within the next five years computer
programs should be able to model semantics, not just simply
mimic
syntax.


By
the year 2029 computers will be completely
indistinguishable from human
intelligence (
Ramaswami
,
2009).

Vernor

Vinge’s

(2008) Scenarios for AI


The AI Scenario: superhuman AI in computers.


The IA Scenario: humans can be essentially upgraded
by computers (Intelligence Amplification).


The Biomedical Scenario: We alter the biological
structures of our brains thus improving human
intelligence


The Internet Scenario: All information becomes directly
connected creating a super intelligent entity.


The Digital Gaia Scenario: Networks of microprocessors
become so efficient that they are said to hold
superhuman intelligence.


Predicting Outcomes


Virtual reality worlds (
Kurzweil
, 2005)


Cures for everything and every possible
discovery made within years of intelligence
explosion (Good, 1965).


If something can go wrong, it will (
Vinge
,
1993).


Post biological beings (Dick
, 2008).

http://www.cyberpunkrevi ew.com/images/Termi nator3
-
09.jpg

http://i mg2.ti mei nc.net/ew/dynami c/imgs/070821/jetsons_l
.jpg

http://www.yugatech.com/ri ngtones/wp
-
content/upl oads/2007/06/transformers
-
movi e.jpg

Which Future Will It Be?

Regulation for AI must start now


The
Singularity Institute for Artificial
Intelligence
Mission Statement


“In the
coming decades, humanity will likely create a
powerful artificial intelligence. SIAI exists to
handle this urgent challenge, both the
opportunity and the risk (singinst.org).”


“Will the movie we live in be The
Jetsons

or
Terminator 2
? (
Stix
, 2000)”.




YouTube


The Singularity

http://www.cyberpunkrevi ew.com/images/i
-
robot07.jpg

References



Bear, M.F., Connors, B.W., &
Paradi so
, M.A. (2007).
Neuroscience: Exploring the Brain
.
Bal ti more
: Li ppi ncott Wi l l iams & Wi l kins
.


Bennett, M.R. and Hacker, P.M.S. (2005). Emoti on and corti cal
-
subcortical

functi on:
conceptual devel opments
.
Progress in
Neurobiology
, 75, 29
-
52
.


Castel vecchi
, D. (August 30, 2008).
Building ‘The Matrix’
. Retri eved February 2, 2009 from




http://sci encenews.org/vi ew/generi c/id/34497/ti tl e/Bui l di ng_The_Matri x
.


Ceurstemont
, S. (2009).
Sociable robots learn to get along with humans.

Retri eved February 23,
2009
from

http
://
www.newsci enti st.com/arti cl e/dn16644
-
vi deo
-
soci able
-
robots
-
learn
-
to
-
get
-
al ong
-
wi th
-
humans.html?DCMP=OTC
-
rss&nsref=onl ine
-
news



Di ck, S. (August 19, 2008).
Could Robot Aliens Exist?
Retri eved February 2, 2009
from

http
://
www.popsci.com/popsci
-
staff/arti cl e/2008
-
08/could
-
robot
-

al i ens
-
exi st


Good
, I. J. (1965). Specul ati ons Concerni ng the Fi rst Ul trai ntel ligent Machi ne
. Advances in



Computers (Academic Press), 6,
31
-
88
.


Hayl es
, N.K. (1999).
How We Became
Posthuman
: Virtual Bodies in Cybernetics, Literature,
and
Informatics.

Chi cago: The Uni versi ty of

Chi cago
Press
.


Kurzwei l
, R. (2005)
The Singularity is Near.
Avai l able February 3, 2008 from http://books.googl e.com/


books?hl
=
en&i d
=88U6hdUi 6D0C&dq=
technol ogi cal +singularity&printsec
=
frontcover&source
=
web&ots
=
v_fYmFuxKI&si g
=9XhPpi m2dKgwi p13jZPGtl 4VQN0
&sa=
X&oi
=
book_resul t&resnum
=3&ct=resul t#PPA39,M1


Moravec
, H. (1988).
Mind Children.
Boston: Harvard Uni versi ty Press.


Natural Moti on

Ltd. (2008).
FAQ.

Retri eved February 23, 2009 from
http
://www.natural moti on.com/faq.htm


Pi nker, S. (1997).
How the Mind Works.
Retri eved February 3, 2009 from
http
://human
-
nature.com/books/mi nd.html
.


Ramaswami
, R. (2009).
Artificial Intelligence: Is the Future Now For A.I.?

Retri eved
February
23, 2009 from

http
://
www.thejournal.com/arti cl es/23927_5


Sti x
, G. (2000).
Artificial Intelligentsia.
Retri eved February 3, 2009
from

http
://www.sci am.com/arti cl e.cfm?id=arti fi cial
-
intel ligentsi a
.


Thomas
, A. (2008).
Cray goes with Intel for HPC
. Retri eved February 16, 2009
from

http
://www.thei nqui rer.net/inqui rer/news/530/1026530/cray
-
goes
-
i ntel
-
hpc
.


Turi ng, A.M. (1950). Computer Machi nery and Intel l i gence.
Mind
, 59, 433
-
460.


Twi st, J. (2005).
Law that has driven digital life.

Retri eved February 16, 2009 from



http
://
news.bbc.co.uk/2/hi/sci ence/nature/4449711.stm


Vi nge
, V. (2008).
Signs of the Singularity .
Retri eved March 1, 2009 from http://www.spectrum.i eee.org/jun08/6306


Vi nge
, V. (1993).
The Coming Technological Singularity: How to Survive in the Post
-
Human



Era
. Retri eved February 2, 2009 from
http://www
-
rohan.sdsu.edu/facul ty/vi nge/

mi sc/si ngularity.html
.