IEEE Technology and Society Magazine - December 2016 - 61

concerned with what we can accomplish, we
neglect to consider what may be the result.
The very same Effect was well illustrated in the
movie Jurassic Park. In speaking of the complete
lack of humility before nature, one of the characters - Ian Malcolm (played by Jeff Goldblum) states
that the: "... scientists were so preoccupied with
whether or not they could [bring back dinosaurs],
that they didn't stop to think if they should" [5].
If we ought to think about the development of AI
consciousness and the potential for self-preservation that may lead to aggressive resistance, do we
need to consider the possibility of what was referred
to in Jurassic Park as a Lysine Contingency1 (i.e.,
a kill switch) in the eventuality of a worse-casescenario? The worst-case scenario or Frankenstein
Effect in this case would involve the unpredictable
(perhaps violent) ways in which conscious AI may
react to the attempt to limit or bring about the cessation of its functions.
Since we have no idea how such an emergent property such as consciousness could affect the behavior
of the AI, it follows that there is a possibility that
incorporating Asimov's Three Laws of Robotics (see
Figure 6), will have little to no effect on preventing the
AI from initiating self-preservation. Without accurately predicting an AI value system, we have little to
no idea how it may react when attaining a state of
self-awareness. Its understanding of itself will be the
result of a manifestation of a myriad number of
sources within its database(s), the algorithmic principles upon which it is relying, etc. We have no idea
how this form of AI could react. Therefore, it is not
inconsistent to maintain that such a form of conscious AI has the potential to usurp any or all of
Asimov's three laws. It is conceivable that AI will not
abide by these laws if it places higher value on its
own existence or the continuation thereof, whatever
way it defines value. We are now faced with the question of considering what safeguards should be in
place before humanity creates such a potential being.
9) The need for an international regulative body on
the development of AI. Due to the fact that AI is
developing at such a rate, it would be prudent to
develop an international body that could act to regu1
From Jurassic Wiki: "The Lysine Contingency was a genetic alteration
Henry Wu performed in the dinosaur genome. The modification knocked
out the ability of the dinosaurs to produce the amino acid Lysine. This
forced the dinosaurs to depend on lysine supplements provided by the
park's veterinary staff. In this way dinosaurs could never escape from the
park because they would never survive long without the food supplements.
It was also explained that without the Lysine, the animals will collapse into
a coma and eventually die. However, it is shown in The Lost World: Jurassic
Park and Jurassic Park III that the dinosaurs can survive well without the
crew giving them Lysine by eating Lysine rich food, like chickens or soya
beans and carnivores could acquire Lysine by eating herbivores" [6].

DECEmbEr 2016

∕

Laws of Robotics
1) A robot may not injure a human being or, through inaction,
allow a human being to come to harm.
2) A robot must obey the orders given to it by human beings,
except where such orders would conflict with the First Law.
3) A robot must protect its own existence as long as such protection
does not conflict with the First or Second Law [7].
Figure 6. Asinmov's Three Laws of Robotics.

late the development of such systems. Some questions
that stand before us are, who precisely should be on
such committees or boards? And why? This will involve
international cooperation, transparency, and trust.
Hopefully, we will collectively act on such an initiative.

Potentially Catastrophic Effects
With rapid advances in development of AI systems
worldwide, it is increasingly prudent and relevant to
consider the issues discussed here. To ignore these
could lead to situations for which we are left unaware
and unprepared, with potentially catastrophic effects.
The next decade will be crucial for the development
of policies, standards, and safeguards to assure that
discovery, invention, and innovation in AI is possible,
but to at the same time allow for security and assurances in anticipation of all potential contingencies.

Author Information
Christopher DiCarlo is a professor of philosophy of science and ethics in the Human Biology Program and Philosophy Department at the University of Toronto,
Toronto, Ontario, Canada. Email: christopher.dicarlo@
utoronto.ca.

References
[1] M. Sainato, "Stephen Hawking, Elon Musk, and Bill Gates warn
about artificial intelligence," Observer, Aug. 9, 2015; http://observer
.com/2015/08/stephen-hawking-elon-musk-and-bill-gates-warnabout-artificial-intelligence/.
[2] Lectures and Essays by the. late William Kindon Clifford, F.R.S., ed.
Leslie Stephen and Frederick Pollock, with an Introduction by F. Pollock, 2nd
Edition (London: Macmillan and Co., 1886), p.344.
[3] C. DiCarlo, "How Problem Solving and Neurotransmission in the
Upper Paleolithic led to the Emergence and Maintenance of Memetic Equilibrium in Contemporary World Religions in "Bioculture:
Evolutionary Cultural Studies," a special evolutionary issue of Politics and Culture, Issue 1, April.
[4]"Ex Machina: A Movie Review" by Christopher DiCarlo, in
Humanist Perspectives, Issue 194, Autumn, 2015. http://www
.humanistperspectives.org/issue194/dicarlo.html
[5] "Jurassic Park (film)," Wikiquote, Oct. 3, 2016; https://
en.wikiquote.org/wiki/Jurassic_Park_(film).
[6] "Lysine contingency," Jurassic Wiki, Oct. 12, 2016; http://
jurassicpark.wikia.com/wiki/Lysine_contingency.
[7] G. Dvorsky, "Why Asimovs Three Laws of Robotics can't protect
us," Gizmodo, Mar. 28, 2014; http://io9.com/why-asimovs-threelaws-of-robotics-cant-protect-us-1553665410.

IEEE TEchnology and SocIETy MagazInE

http://observer http://www http://www.humanistperspectives.org/issue194/dicarlo.html http://https:// http://http:// http://jurassicpark.wikia.com/wiki/Lysine_contingency http://www.io9.com/why-asimovs-three

Table of Contents for the Digital Edition of IEEE Technology and Society Magazine - December 2016