Hetvi Parekh
Intelligence of Artificial Intelligence: Philosophy
Abstract
Artificial intelligence (AI) has been rapidly spreading over the human world like the spread of
any virus. Artificial intelligence plays its game from making a simple web search - to driving
a car - to making a machine acting like a human (robot) - to launching a shuttle into the space.
In such scenario, the question arises that whether artificial intelligence is completely
duplication of human intelligence. To answer this question, philosophy highlights an important
aspect of ‘consciousness’, which artificial intelligence is deprived of. Also, artificial
intelligence and pre-programmed machines somewhere neglect the virtues and morals that are
imbibed within the humans. To fill up such voids in the working conditions, human
intervention becomes undeniable. Various philosophical arguments, ethical problem theories
and scientific, technological and fictional experiments blend in as evidences to draw the
superiority among the human intelligence and artificial intelligence.
Keywords
Artificial Intelligence, AI, Machines, Human Intelligence, Consciousness, Philosophical AI
___________________________________________________________________________
Introduction
You might have mentioned to a friend of yours while chatting that you were looking for some
particular shoes and the next moment, while opening a browser or some app, you’ll notice the
advertisement of the same shoes. You’ll be tempted to visit that link and you may even end up
ordering them. But how did your browser or that app come to know that you were interested in
buying those shoes? The answer is ‘Artificial Intelligence’, popularly known as ‘AI’.
Technology historian George Dyson rightly puts it, “Are we searching the search engines or
are the search engines searching us?”. In video games, for instance, in the game of Chess,
how does the computer plan its move like a human being? Again, it’s because of artificial
intelligence. You may often wonder that your Siri or Google assistant is so intelligent, but that
intelligence is completely artificial. The driverless cars like Tesla seem to be so magical. But,
the magician driving the car is an artificial intelligence. Also, robots – the enchanting creations!
These robots have a body like humans, they act like humans and they work like humans; but
are just not humans. They are the product of artificial intelligence. It seems that artificial
intelligence has started taking place of humans in almost every field. Human intelligence has
indeed got a competitor!
1/5 | P a g e
Intelligence of Artificial Intelligence: Philosophy
Artificial Intelligence (AI) and its Working
Artificial intelligence (AI) is the field devoted to building artificial animals (or at least artificial
creatures that – in suitable contexts – appear to be animals) and, for many, artificial persons
(or at least artificial creatures that – in suitable contexts – appear to be persons).1 The
prominent-possible goals of AI are to develop systems that think rationally, systems that think
like humans, systems that act rationally and systems that act like humans.2 There are basically
four types of artificial intelligence: reactive machines, limited memory, theory of mind and
self-awareness.3 In these four types, almost all possible forms, simulating to human intelligence
are covered. Explosive growth of AI has led to bloom in machine learning4, the resurgence of
neurocomputational techniques5 and the resurgence of probabilistic techniques6. We are
generally aware that machines are pre-programmed and follow the codes to act, which is quite
intelligent. But, even more complex paradigms persist in the reformed versions of artificial
intelligence. To put it up in laymen expression, in such versions, objects powered by artificial
intelligence draw their knowledge by acting like sensors that study the present environment,
evolution in the environment, the result of outcomes given by them and then, decide what
action they should undertake, which is again directed by some action-condition rules. 7 As it is
basically about gaining knowledge, artificial intelligence has a strong connection with the
philosophical branch ‘Epistemology’. Hence, the role of philosophers has remained vital in the
development of AI.
Deprivation of Consciousness
Looking at the pace of development of artificial intelligence, it seems that it holds the capacity
to indistinguishably replicate human intelligence and create artificial human minds someday.
But, there is something, existence of which looks almost next to impossible in artificial
intelligence; atleast as of now. It is our human ‘consciousness’. Various philosophical theories
and arguments readily prove that machines are incapable of having conscious intelligence.
Argument I
“I think, therefore, I am.” – Rene Descartes.8 Descartes in one of his mediations, starts
doubting everything, including the existence of his own self. From this method of skepticism,
he conceives that since he has the ability to doubt and think about his own self, he surely exists.9
This prominent philosopher of history, Descartes who gave birth to ‘mind and body dualism’,
also says that, “Self is that which doubts, understands, affirms, denies, is willing, and also
imagines and has sensory perceptions”. Descartes considered mind and body as completely
different entities. He considered our body to be a machine and mind to be a conscious substance
that drives the bodily machine. To sum up, he declared that if an entity has a capacity to doubt
and think, only then it has consciousness. Further, only that entity which has consciousness,
can be a mind and rest, all the things are just machines.10 In case of computers and machines,
they do not have the ability to doubt their existence and thus, cannot think about their self and
therefore, they cannot have a conscious mind like humans.
2/5 | P a g e
Intelligence of Artificial Intelligence: Philosophy
Argument II
Machines designed so far are totally deprived of consciousness. Then why are we sometimes
mistaken that they are answering to us as if they know everything? It is because they carry out
actions, but actually aren’t aware of the doings. “Machines do not have consciousness, mind
and sentience but only simulate thought and understanding”, says famous philosopher John R.
Searle.11 It can be clearly understood by Searle’s Chinese Room Argument. John is locked
inside a room with a box of codes in Chinese as well as in English and there are Chinese men
outside the room that pose questions to John in Chinese and John has to answer those questions,
but, John is not acquainted even with a single word of Chinese. John successfully answers all
the questions posed to him in Chinese with the help of the code box, inspite of not knowing
even a single word of Chinese. Just as a computer can answer to us in many languages inspite
of not knowing any language; they aren’t conscious. 12 Machines are programmed with such
codes and with the repetitive actions, they master an action without even knowing what is
happening, just like John in the experiment. They deliver the work to us as if they have a
working mind, but in reality, they have bits and pieces of information, they connect them
together, observe the user interaction and put the output onto the user’s plate.
Argument III
To specifically talk about the actual process of thinking just like humans, AI is further divided
into ‘weak AI’ – that simulates thinking and ‘strong AI’ – that actually has ability to think like
humans.13 Philosophers and technologists have referred to weak AI as easily achievable, but
the concept of strong AI has been very controversial since years, because strong AI exclaims
the presence of consciousness. American mathematician, Kurt Godel presented a theory called
‘theory of incompleteness’ as a sharp attack on the strong AI. As per this theory, Godel says
that there are some propositions that do not have proof or have incomplete proof. Since, they
do not have proof, they are regarded as false, but are actually true.14 We as conscious beings
can conceive those prepositions to be true even without proof with our common sense,
emotions and/or sometimes even with our gut feelings, whereas, in such a case, machines will
display an ‘error’. After analysing Godel’s theory of incompleteness, philosopher J. R .Lucas,
proposes a regress that for a self to be conscious, it should be aware and it should be aware that
it is aware and further, it should be aware that it is aware and should be aware about it and it
goes on. Further, an entity should be aware about this as one and not in parts just as humans
are aware about it just by mind.15 Thus, there are characteristics for consciousness that cannot
be induced within the machines, even though formalised completely.
Argument IV
From the above arguments, it seems that yet and in the near future, artificial intelligence will
remain deprived of consciousness. But, there’s another serious problem with artificial
intelligence due to lack of consciousness, which is that it cannot have morals, values, ethics,
emotions, feelings and common sense adhering to religion, family background, nationality,
culture, mythology, education, etc., which a human develops within herself as it grows.
3/5 | P a g e
Intelligence of Artificial Intelligence: Philosophy
Philosopher Phillipa Foot has developed a problem called ‘Trolley Dilemma’ on the same
ethical lines. The problem says that a serious dilemma arises when a runaway trolley is moving
on the road and it is sure that the death of person/persons on the either side of the road is
inevitable due to the accident. The trolley can’t be stopped and the switch to divert the trolley
to either of the roads is in the hands of humans. This problem is mainly related to the automatic
driverless cars.16 There is no particular solution to this problem, but humans with the above
stated characteristics that they inherit can make an on-the-spot rational decision. But for
machines, such decision is seemingly impossible. There are high chances that a driverless car
might make an unlikely decision with rigorous consequences.
Conclusion
Considering the various discussed grounds, it seems that despite the unfathomable efforts in
the development of artificial intelligence, atleast as of now, it is next to impossible that it can
completely duplicate human intelligence. It is unceasingly intelligent, but at the same time, it
is dumb without consciousness. The field of artificial intelligence is the product of superhuman
intelligence and there can be no denial in the fact that proper development in this field will lead
to wonders and make life of humans much easier. Artificial intelligence may or may not
duplicate human intelligence completely, but it sure to equip humans with tremendous
superpowers, making the world magical. It is a wonderful multi-disciplinary field that requires
inputs of philosophers, engineers, programmers, ethicists, technicians, mathematicians,
psychologists, data scientists, biologists and many more. Now, only time can decide where the
development takes us. We cannot completely eliminate the idea that in 22nd century, there
would be robots in the crowd of humans, totally indistinguishable and then, we would be
struggling over problems like how to maintain a peaceful nexus between human minds and
artificial minds.
References
1
Bringsjord, S. and Govindarajulu, N. (2018, July 12). Retrieved from Stanford
Encyclopedia of Philosophy: https://plato.stanford.edu/entries/artificial-intelligence/
2
Bringsjord, S. and Govindarajulu, N. (2018, July 12). Retrieved from Standford
Encyclopedia of Philosophy:
https://plato.stanford.edu/entries/artificial-intelligence/#WhatExacAI
3
Hintze, A. (2016, November 14). Retrieved from Government Technology:
https://www.govtech.com/computing/Understanding-the-Four-Types-of-Artificial-
Intelligence.html
4
Bringsjord, S. and Govindarajulu, N. (2018, July 12). Retrieved from Standford
Encyclopedia of Philosophy: https://plato.stanford.edu/entries/artificial-
intelligence/#BlooMachLear
4/5 | P a g e
Intelligence of Artificial Intelligence: Philosophy
5
Bringsjord, S. and Govindarajulu, N. (2018, July 12). Retrieved from Stanford
Encyclopedia of Philosophy: https://plato.stanford.edu/entries/artificial-
intelligence/#ResuNeurTech
6
Bringsjord, S. and Govindarajulu, N. (2018, July 12). Retrieved from Stanford
Encyclopedia of Philosophy: https://plato.stanford.edu/entries/artificial-
intelligence/#ResuProbTech
7
Bringsjord, S. and Govindarajulu, N. (2018, July 12). Retrieved from Stanford
Encyclopedia of Philosophy: https://plato.stanford.edu/entries/artificial-
intelligence/#InteAgenCont
8
(1644). In R. Descartes, Principles of Philosophy (Part 1) (p. article 7)
9
(n.d.). Retrieved from Spark Notes:
https://www.sparknotes.com/philosophy/meditations/section3/
10
Nath, R. (2010, November). Retrieved from Research Gate:
https://www.researchgate.net/publication/228369537_A_Cartesian_critique_of_the_ar
tificial_intelligence
11
(1980). In J. Searle, Mind, Brain and Programs.
12
Cole, D. (2020, February 20). Retrieved from Stanford Encyclopedia of Philosophy:
https://plato.stanford.edu/entries/chinese-room/#ChinRoomArgu
13
(2019, May 27). Retrieved from E-3 Magazine International:
https://e3zine.com/strong-artificial-intelligence/
14
Sautoy, M. (2017, May 31). Numberphile. Retrieved from Youtube:
https://www.youtube.com/watch?v=O4ndIDcDSGc&t=144s
15
Lucas, J. R. (n.d.). Mind, Machines and Godel, 1961. Retrieved from Jstor:
https://www.jstor.org/stable/3749270?read-
now=1&refreqid=excelsior%3A77afa50d0c3b8e49d2c847d5e1e278c0&seq=1
16
(2019, February 11). Retrieved from StrongBytes:
https://strongbytes.ai/the-trolley-dilemma-in-ai/
___________________________________________________________________________
Hetvi Parekh | Ahmedabad University | 2020, April 24th
5/5 | P a g e
