Artificial intelligence and ethics

  Abstract

Throughout the next essay, several questions about artificial intelligence will be raised. Arbitrarily, some questions will be answered and other will only be raised. As you read, you will see a great influence of science fiction, of a world complemented by the AI, with automatas, independent operating systems and a lot of new circumstances to be solved on issues of identity, security and ethics.

Source: https://www.instagram.com/p/Be6VZEzgAEk/?taken-by=elonmusk

Introduction

The constant evolution of technology and science has resulted in consequences directly proportional to the development of societies. The machines are part of the technological evolution, and have managed to help with different types of jobs, at the same time that we talk about intelligent machines, we usually talk about term automata (αὐτόματα), to refer to mechanical mechanics properly spontaneous that act by themselves , (Language, 2018). Therefore, approximately from the third century B.C. (Geographic, 2018), several questions have been proposed mathematically and philosophically about the true spontaneity of a system and its complex repercussions.

The formally called artificial intelligence (AI), has been described on multiple occasions by scientists and science disseminators such as Stephen Hawking, Isaac Asimov, Alan Turing and Roger Penrose, in novels, treatises, or books. Currently it is known that there are two types of AI; strong or similar to human, and weak or applied, such as Siri of iOS, (Goertzel, 2014).

For to open the way to the most common questions, an example of an analogy is mentioned: A father teaches and helps develop their children, these usually have the tendency to become more intelligent, capable or strong than their predecessors, (Campbell & Reece, 2007, pág. 513). In our species it is common to see how the parents, after certain years, look at their children being much more successful than him at their age, and they feel proud, biologically, for having transcended the next generation, and intellectually for having given them the path or the efficient mental structures so that they develop, obtain more benefits and in turn can be happy. As an analogy it is understood that the machines are a creation of the human intellect, and consequently their daughters. Would we be proud to be replaced by the efficiency of AI? (Penrose, 1989, pág. 30), What measure would we have for the preservation of our species through DNA? In case of restricting that our creations learn and choose at their own will what to do with their time (AI applied), What laws should be implemented in its system?  Perfection is a utopian concept, therefore, its systems are susceptible to corruption or hacking, What protocols should be taken in case they are granted a free act or will of their own?, Automatic vigilance without limits ?, Zero privacy?, Who would control ?, Could AI commit crimes?

Security

Hundreds of questions about security have been hypothesized because of the continued progress of intelligent systems, the same ones that are currently occupied worldwide as Android, iOS, Windows, Linux, and others.  These are interconnected immediately through a connection to the Internet, with "tons" of information, being a potential tool, with powerful scope to solve and create problems. However, civilization cannot be stuck in fear without foundations, so, it is necessary to have plans in risk management, work on physical laws and programming code so that technology is safe. As exemplified, Asimov in one of his books, (I, Robot, 1950): 

“The Three 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 order given 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 Laws.

Handbook of Robotics

56th Edition, 2058 A.D. “

Although several hypothetical questions could be products of a post-apocalyptic paranoia influenced by science fiction for some years, (Wiseman & Fahey, 2013), humans have seen how various things in science fiction have become reality. The machines are programmed through orders written in programming code that are governed by mathematical algorithms previously designed for a specific purpose.

In theory any grammatical statement is executable. In theory any grammatical statement is executable, for example: calculate all income, get the data traffic of the month, bring a coffee, drive to my girlfriend's house, monitor my pets, protect my family. In the same way it can be observed that it can be simple to increase a "no" before each statement and this would not be nice, however, in programming language, these statements should have enough security concatenated, in such a way, that alarms, in case of corruption attempt in the code and thanks to engineering there are already many security protocols that can be used.

Seen in a positive security perspective, AI can offer quantum levels of security, for different purposes, including against terrorism, (Agency, 2018, pág. 12), the same that can be implanted in different electronic devices, such as facial recognition on iPhone smartphones, (Inc, 2017). The applications of AI are infinite in security fields, an example is in global economy, a quantum computer with an applied AI would be able to prevent financial disasters.

Ethics

One of the benefits of AI is the prevention of global terrorism through automatic face recognition, voice recognition, or certain patterns, however, the principle of these algorithms is to work stealthily. Is it ethical to violate personal privacy? To what extent is a world without secrets beneficial?

Science is not far from achieving a AI strong in operating systems, with the new quantum computers in a few decades, "high" artificial intelligence, could be able to create art, do science and feel empathy or fear. What kinds of relationships or mental connections would be established between a strong AI and a human? Different directors or writers have shown different possible interactions between the human and a strong AI in films like Her, Halo, Ex machine, Bicentennial Man, and in particular you can extract certain questions. How long would a human be interesting for an AI? Would it be real attraction? A kind of experimentation?

The "superior intelligence" of the human or biological programming (altruism, empathy, fear, love, anger, interest, instinct of conservation, etc.), has allowed to create strong beneficial links, and an example of this is the relationship of the human with other animals such as dogs, cats, cows, birds, and plants, where you can see the existence of a successful symbiosis beneficial for several. Or it may end up being beneficial only to one as in the case of humans and cattle.

Philosophy robotic

The identity is a conceptual construction that allows to point out an equality or belonging and at the same time highlight its uniqueness. To understand the concept of identity well, it is indicated its origin from latin idem which means "equal" or "the same", (Cortina, 2006, pág. 61). In philosophical terms, identity is a characteristic of the entity (ens), Therefore, it is considered as the participle of the verb "to be", a concept that can be carried to different grammatical modes and times. This means that, in reality, an entity potentially: is, is not, was, or could be, (Copesa, 2009). 

An example is the case of Sophia, “who” has received nationality, and learns different human gestures (characteristic of intelligence), (Robotics, 2017), and it's amazing, since the nationalities were concepts part of the just human identity, in fact the dogs (Canis lupus familiaris), "the best friend of the man", they do not have nationality although they take approximately 9000 years very close to the man, (Axelsson & Lindblad-Toh, 2012).

For the principle of identity of the Aristotelian philosophers “ents est ents”, like "I am who I am" or "it is what it is", it refers to identity as the assimilation of a concrete reality in a definite time, (Cortina, 2006, pág. 62). For example, a human is a human, the sun is the sun, and a robot is a robot, all these at a definite time of the present. What happens if a human is implanted with many robotic parts, so that they occupy more than fifty percent of their total mass?, Is the person being intervened still a human being? With the same thread of supposition, if instead of receiving many robotic implants, only receive a robotic brain implant that involves mental abilities, such as recovering memory, improving it, or a software capable of giving information of any kind, Would this individual still be human? The singularity of convergence between the known mental capacities of a human and the mental capacities would create a new identity, and perhaps "biologically" a new species.

Identity problems in humans can be very important. As a set of mental constructs, the events in a person's life, can be positive or negative for their lives and for others, and these are defining their personality. Would the evolution process of the identity of an AI be more advanced than that of a human? Thanks to this evolution of individual identity can answer several questions like Who am I ?, What I like ?, Why do I do this ?, Am I good, bad, what is that ?, Although many of these philosophical questions are not very simple, many can be answered over time.

Surely there are more interesting questions and answers, however, there are several beliefs within the identity that can act in determined ways very established, making a certain behavior. For example, in the competitions of the Olympic Games, the national anthem of each one of the competitors is usually placed at the top, at that moment the spectators with the same nationality who feel identified, stand up, or sing only the anthem of your country like all your countrymen, (Fernández, 2016, pág. 122); another example may be religion, many people acquire a belief system of some religion by inheritance or other circumstances and these make them act in a certain way, and in a certain way some can feel identified, seeing a cross, others seeing a star, a moon, a sun, a number, or seeing a book by Leibnitz, (Guardans & Puigardeu, 2009, pág. 10). 

What beliefs would an AI adopt with all the knowledge that currently exists on the internet, in databases, books, stories of people, among other documented knowledge? At what time could you establish a relatively stable personality? What attitude and behaviors would a machine with AI take? If the identity is directly proportional to the assimilated, analyzed and accepted information, a mathematical model could be made that involves a function of the discoveries of reality as a function of time and a function of the information received by an AI.

For example, if the reality had a mathematical function that tends to a specific value, it is considered that the AI could arrive at a behavior with stable character, and on the other hand, if the representation of the "reality" had a form similar to an exponential function for the domain of all positive and negative real numbers, it is expected that the evolution of identity and mental processes also have that direction. With the assumption (two dimensions x, y) that an AI has information without restrictions or deprivation, that the reality is infinite as a function x ^ 3, (Stewart, 2012, pág. 49), in the whole domain of the real, it follows that the evolution of the identity of the AI would have two opposite directions. It is possible to create successful mathematical models, maybe create an AI, in any case time will tell the evolution of the human species and its creations

References 

Agency, N. S. (2018, January). www.nsa.gov. Retrieved from https://www.nsa.gov/resources/everyone/csfc/capability-packages/assets/files/dar-cp.pdf

Asimov, I. (1950). I, Robot. California: Universidad de California - Dutton.

Axelsson, E., & Lindblad-Toh, K. (2012, 12 11). The genomic signature of dog domestication reveals adaptation to a starch-rich diet. Nature international journal of science. Retrieved from https://www.nature.com/articles/nature11837

Campbell, N. A., & Reece, J. B. (2007). Biología (7ma Edición).[Biology] Buenos Aires: ES: Médica Panamericana .

Copesa, G. (2009). Icarito. Retrieved from http://www.icarito.cl/2009/12/97-4292-9-4-verbos.shtml/

Cortina, I. J. (2006). Identidad, identificación, imagen.[Identidy,indentification,image]. México D.F.: Fondo de Cultura Económica.

Fernández, P. E. (2016). Juegos olímpicos, televisión y redes sociales.[Olympic games, television and social network]. Barcelona: Editorial UOC. Retrieved from https://ebookcentral.proquest.com

Geographic, N. (2018, 01 05). NAT. Retrieved from http://www.nationalgeographic.com.es/historia/grandes-reportajes/inventos-griegos_9395/6

Goertzel, T. (2014). The path to more general artificial intelligence. Journal of Experimental & Theoretical Artificial Intelligence Vol 26, 343-354. Retrieved from https://doi.org.ezbiblio.usfq.edu.ec/10.1080/0952813X.2014.895106

Guardans, T., & Puigardeu, Ó. (2009). Una historia de las religiones.[A history of religions]. Barcelona: Ediciones Octaedro. Retrieved from https://ebookcentral.proquest.com

Inc, A. (2017, 12 20). support.apple. Retrieved from https://support.apple.com/en-gb/HT208108

Language, R. A. (2018, 01 5). RAE. Retrieved from http://dle.rae.es/?id=4TKy9Vs

Penrose, R. (1989). The Emperor's New Mind. Concerning Computers, Minds, and The Laws of Physics. Retrieved from https://libreepub.org

Robotics, H. (2017). http://sophiabot.com. Retrieved from http://sophiabot.com/about-me/

Stewart, J. (2012). Cálculo de una variable trascendentes tempranas 7ed.[Calculus, Early tracedentals]. México D.F.: Cengage Learning Editores, S.A.

Wiseman, P. C., & Fahey, J. (2013). Practically human: Can smart machines do your job? New York: Associated Press: The Big Story.