oh, karlo

Using Rawls' concepts in Justice as Fairness as solutions for a possible technological unemployment and underemployment due to job automation brought about by artificial intelligence

14 min read

Originally submitted as a requirement for the course Philo 1 under Alexander Atrio Lopez.

September 2018

Introduction

Groover (2016) defines automation as the technology by which a process or procedure is performed without human assistance.[1] Automation or automatic control is also defined as the use of various control systems for operating equipment such as machinery, processes in factories. (Rifkin, 1995)[2] The history of automation goes a long way back, from around 270 BC when Ctesibius devised a number of machines, including an improved water clock equipped with what is now believed to be the first feedback control system.[3]

In the 18th century Europe, Thomas Newcomen invented the modern steam engine which marked the start of the Industrial Revolution that brought about the mass production of machineries to hasten processes created by manual labor.[4] During the World War II, British cryptanalyst and computer scientist Alan Turing developed a computer which could break German ciphers much faster than a group of mathematicians could[5], but while it helped them win the war, it in turn eliminated the need for manual computers (a term for people who did computing as a profession) and displaced them from jobs. This phenomenon is called technological unemployment. As an example, in 1920, there were 1.2 million coal miners in the United Kingdom. In 2012, the number employed was less than 5000.[7]

With the advent of new technology, a relatively new term Artificial Intelligence is rapidly making its way in our society. Dragonette (2018) describes Artificial intelligence (AI) as “a term for simulated intelligence in machines. These machines are programmed to “think” like a human and mimic the way a person acts.” From self-driving cars and self-flying planes to robots that can cook three-hundred pizzas in an hour, AI is integrating into our lives and workplaces, performing tasks alongside us and possibly in the future completely take over our jobs.[8]

In a research conducted by McKinsey & Company, an estimated 800 million jobs worldwide could be automated by 2030. In the United States, a survey by Pew Research Center on 4,135 U.S. adults shown that 72% are worried about robots taking over their job while only 33% are enthusiastic about it.[9] In an interview by AJ+, an online news channel by Al Jazeera Media Network, Trucker Brown, a truck driver and an online personality, expressed his concern on self-operating trucks, especially because millions of truck drivers only have driving as their means of generating income for livelihood of their families.[8] U.S. White House Council of Economic Advisers in 2016 estimated a total of 3.1 million jobs from the car industry (taxis, delivery trucks, chauffeurs, etc.) to be at risk from job automation.[10] This is excluding all the other jobs at risk of automation such as in the medical fields, where computers can read medical scans about 10,000 times faster than even humans qualified for the job can.

Philippine Statistics Authority in 2017 categorized 39.3 million employed persons into three sectors: Services making up the largest group at 57.1% of the total population, followed by Agriculture at 25.5%, and lastly, Industry at 17.4%[11], all of which have an automation potential of at least 48%[12], meaning at least 19 million people are at risk of being displaced by automation. According to a news report by ABS-CBN News, AI is currently replacing call center agents in the business process outsourcing (BPO) industry, cutting around 90% of calls made to human workers.[13]

In this paper, we will argue how three concepts in John Rawls’ Justice as Fairness can provide solutions to the problem of an impending possibility of displacement of humans from their jobs due to automation caused by Artificial Intelligence, leading to unemployment and underemployment: (1) A feature in Rawls’ first principle: that the basic rights and liberties must not be traded off against other social goods, (2) The first part of the second principle: fair equality of opportunity, (3) The original position and the veil of ignorance. This paper will discuss only the aforementioned concepts and nothing more. For (3), the paper assumes that an Artificial Intelligence in the future will evolve into a sentient being capable of intelligence, emotions, and senses.

1. Basic Rights and Liberties against other Social Goods

Wenar (2008) provides an example of how basic rights and liberties must not be traded off against other social goods.[14] To quote:

“The first principle disallows, for instance, a policy that would give draft exemptions to college students on the grounds that educated civilians will increase economic productivity. The draft is a drastic infringement on basic liberties, and if a draft is implemented then all who are able to serve must be equally subject to it, even if this means slower growth. Citizens’ equal liberty must have priority over economic policy.”

In his example, Wenar says that giving exemptions to college student for military drafting would violate the Justice as Fairness’ first principle which states that all citizens should have the familiar basic rights and liberties. Although it may lead to an increase in economic productivity, it would be unfair to those who are not given the exemption. The same principle can be applied to technological unemployment. While labor productivity may increase due to automation[7], due to a number of factors such as that machines can work 24/7 and faster production time, a consequence would be a lot of displaced human workers.

A solution to this problem using Rawls’ concept would be to regulate the work allotted to machineries, robots, and artificial intelligence. By regulate, we mean to balance the work allotment between humans and machines assuming that automation only has rendered the human workers working part-time. If possible, maintain the working hours for humans as it was before the occurrence of automation for the specific job. For example, move the machine’s working hours such that it would only take up idle time shifts, meaning the time when a human worker is not available. Also, these machines may work alongside humans for an increased productivity.

In case automation has taken over the whole job, a possible solution would be relocating the displaced humans to a different work opportunity. This, however, would lead to another problem: underemployment, which is defined by Feldman (1996) as the under-use of a worker due to a job that does not use the worker’s skills.[16] The workers who will be moved to a different job may not be able to perform their work properly simply because they are not qualified to do so.

The most Rawlsian approach to the problem of automation on the first principle then, is to disallow the use of machines altogether for the benefit of the people, as this will maintain their equal liberty best by giving them the most hours to work and jobs to do.

2. Fair Equality of Opportunity

According to Wenar, the first part of the first principle, fair equality of opportunity requires that citizens with the same talents and willingness to use them have the same educational and economic opportunities regardless of whether they were born rich or poor.[14] Quoting from Rawls’ A Theory of Justice:

“In all sectors of society there should be roughly equal prospects of culture and achievement for everyone similarly motivated and endowed. The expectations of those with the same abilities and aspirations should not be affected by their social class.”[17]

As stated earlier, the Philippines’ workforce is composed mainly of three categories services, agriculture, and industry, all of which have automation potential. These categories are further broken down into many sectors, including but not limited to, forestry and fishing, manufacturing, accommodation, food services, education, finance, health, and construction.[12] In a possibility of automation of these jobs in the near future, the people collectively have to decide what course of action to take. Earlier in subtopic 1, we specified three alternative solutions to the problem of automation brought about by AI:

  1. Regulate the work allotment of the machineries, robots, and artificial intelligence by relegating their work only during human idle times or work alongside them.
  2. Relocating the displaced human workers to different jobs, which may be not of their expertise.
  3. Disallow the use of machines, robots, and artificial intelligence altogether for the sake of the job opportunities for the people.

For these three cases, we can have three statements using Rawls’ Fair Equality of Opportunity: (1) fair equality of opportunity exists because the humans and machines work side by side for better production, (2) fair equality of opportunity may be achieved by giving the displaced sectors the opportunity to learn the occupations they will be relocated to by providing them the education and economic support they need, and (3) fair equality of opportunity exists because the human workers will all fully get their jobs back, and they may pick the jobs they want or jobs that they are good in without the hindrance of automation.

3. Original Position and the Veil of Ignorance

In Rawls’ concept of original position, people select the principles that will determine the basic structure of the society they will live in. This choice is made from behind a veil of ignorance, which deprives participants of information about their particular characteristics such as their ethnicity, social status, gender and, their concept of morality. This forces participants to select principles impartially and rationally. As Rawls described veil of ignorance in A Theory of Justice:

“No one knows his place in society, his class position or social status; nor does he know his fortune in the distribution of natural assets and abilities, his intelligence and strength, and the like”[17]

McCarthy (2006) wrote in his article The Philosophy of AI and the AI of Philosophy “Artificial intelligence (AI) has closer scientific connections with philosophy than do other sciences, because AI shares many concepts with philosophy, e.g. action, consciousness, epistemology (what it is sensible to say about the world), and even free will.”[18] The philosophy of artificial intelligence attempts to answer such questions as follows.[19][20]

  1. Can a machine act intelligently? Can it solve any problem that a person would solve by thinking?
  2. Are human intelligence and machine intelligence the same? Is the human brain essentially a computer?
  3. Can a machine have a mind, mental states, and consciousness in the same way that a human being can? Can it feel how things are?

This topic has been a debate going on since more than half a century starting with Turing’s (1950) Turing Test which in its basic form states that “If a machine behaves as intelligently as a human being, then it is as intelligent as a human being.”[21] In a possible future where Artificial Intelligence has evolved to a point that they are sentient such that they have the previously mentioned characteristics in the philosophy of AI, how do the people consider these machines? If a machine has the mental states, emotions, senses, and consciousness a human being has, is it possible to consider it as a human being? If that is the case, then they must belong in the society and be included in the original position. If the people in the society do not consider the machines as human beings, then they are not included in the original position.

We can separate these arguments into two cases:

  1. If Artificial Intelligence has transcended the capabilities of a human being, then it could be possible that they have assumed the occupations which require several human attributes like emotion and understanding of body language, such as doctors and lawyers. In this case, supposing the AIs are only programmed to do morally good, then what risk could it pose if they are integrated in the society? After all, it would be for the benefit of all the beings involved. As in Rawls’ definition of justice: the arrangement of social institutions that is morally best.[14] Therefore, using the veil of ignorance to put the society in the original position including the machines with Artificial Intelligence would be beneficial, as it also solves the problem of these machines taking the job of human workers. Note however, that this would only be possible if the humans in the society agree to consider the machines as sentient beings that can coexist with humans, treated as humans themselves and have human rights.

  2. If the humans in the society do not agree to consider the machines as sentient beings and given the same rights as humans, then it would mean that they are not included in case of a societal reset using the veil of ignorance for the original position, which by then we could refer to the previous solutions mentioned in subtopics 1 and 2 to solve the problem of technological unemployment and underemployment due to automation.

Summary

In summary, the paper introduced the problem of technological unemployment and underemployment due to automation brought about the advancement in technology and the evolution of artificial intelligence, its causes and its implications, and its application to foreign and local settings. We related three concepts from John Rawls’ Justice as Fairness namely: (1) Basic Rights and Liberties against other Social Goods, (2) Fair Equality of Opportunity, (3) Original Position and the Veil of Ignorance; and used them to propose solutions to the problems.

Computer Science majors are essential in solving the problems presented in this paper most importantly because their field of expertise is concerned with the Artificial Intelligence that pose a problem in our today’s society and in the possible future. They are the ones who program the machines which may be in good cause (e.g. to increase productivity and decrease operating costs) but unfortunately lead to inevitable consequences such as the main problem mentioned and analyzed: technological unemployment and underemployment. At the same time, Computer Science majors study the ethics and philosophy of AI which can help them develop machineries with more ethical considerations.

Bibliography

  1. Groover, M. P. (2016). Fundamentals of modern manufacturing: Materials, processes, and systems. Hoboken, NJ: John Wiley & Sons.
  2. Rifkin, Jeremy (1995). The End of Work: The Decline of the Global Labor Force and the Dawn of the Post-Market Era. Putnam Publishing Group. pp. 66, 75. ISBN 0-87477-779-8.
  3. Guarnieri, M. (2010). The Roots of Automation Before Mechatronics [Historical. IEEE Industrial Electronics Magazine, 4(2), 42—43. doi:10.1109/mie.2010.936772
  4. Morris, Charles R. Morris; illustrations by J.E. (2012). The dawn of innovation the first American Industrial Revolution (1st ed.). New York: PublicAffairs. p. 42. ISBN 978-1-61039-049-1.
  5. Beavan, C., & Stacey, N. (Directors). (2011). Britain’s Greatest Codebreaker [Motion picture on TV]. United Kingdom: Channel 4.
  6. Dragonette, L. (2018, April 19). Artificial Intelligence - AI. Retrieved September 19, 2018, from https://www.investopedia.com/terms/a/artificial-intelligence-ai.asp
  7. Pettinger, T. (n.d.). Technological unemployment. Retrieved September 21, 2018, from https://www.economicshelp.org/blog/glossary/technological-unemployment/
  8. AJ+. (2018, January 23). Robots And AI: The Future Is Automated And Every Job Is At Risk [Automation, Pt. 1] [Video file]. Retrieved September 19, 2018, from https://www.youtube.com/watch?v=rnBAdnNIIXk
  9. Smith, A., & Anderson, M. (2018, May 14). Americans and Automation in Everyday Life. Retrieved September 19, 2018, from http://www.pewinternet.org/2017/10/04/automation-in-everyday-life/
  10. Artificial Intelligence, Automation, and the Economy (Rep.). (2016, December). Retrieved from https://www.whitehouse.gov/sites/whitehouse.gov/files/images/EMBARGOED AI Economy Report.pdf
  11. Employment Situation in January 2017 (Final Results). (2017, September 26). Retrieved September 19, 2018, from https://psa.gov.ph/content/employment-situation-january-2017-final-results
  12. Moraje, S. (n.d.). Seizing the automation opportunity in the Philippines (Rep.). Retrieved September 19, 2018, from McKinsey & Company website: https://www.mckinsey.com/~/media/McKinsey/Featured Insights/Asia Pacific/Seizing the automation opportunity in the Philippines/Seizing the Automation Opportunity in the Philippines.ashx
  13. Displaced by robots, PH call center agents shift gears. (2018, February 28). Retrieved from https://news.abs-cbn.com/business/02/28/18/displaced-by-robots-ph-call-center-agents-shift-gears
  14. Wenar, L. (2008, March 25). John Rawls. Retrieved September 21, 2018, from https://plato.stanford.edu/entries/rawls/
  15. Chui, M., Manyika, J., & Miremadi, M. (n.d.).Where machines could replace humans—and where they can’t (yet). Retrieved September 21, 2018, from https://www.mckinsey.com/business-functions/digital-mckinsey/our-insights/where-machines-could-replace-humans-and-where-they-cant-yet
  16. Feldman, D. C. (1996). The nature, antecedents and consequences of underemployment. Journal of Management, 22(3), 385-407. doi:10.1177/014920639602200302
  17. Rawls, John (1999). A Theory of Justice. pp. 63, 118. ISBN 0-674-00078-1
  18. McCarthy, J. (2006, June 25). The Philosophy of AI and the AI of Philosophy (Stanford University, 2006)
  19. Russell, Stuart J.; Norvig, Peter (2003), Artificial Intelligence: A Modern Approach (2nd ed.), Upper Saddle River, New Jersey: Prentice Hall, ISBN 0-13-790395-2
  20. Fearn, Nicholas (2007), The Latest Answers to the Oldest Questions: A Philosophical Adventure with the World’s Greatest Thinkers, New York: Grove Press
  21. Turing, Alan (October 1950), “Computing Machinery and Intelligence”, Mind, LIX (236): 433-460, doi:10.1093/mind/LIX.236.433, ISSN 0026-4423

Graph View