The Impact of AI on the Future of Work: A Philosophical Approach (Part I)
By Leonhard Teichert, Summer Intern
How will artificial intelligence (AI) change the future of work? This certainly is one of the most relevant and important issues implied by current technological development. Especially because the development of AI does not only affect a far off, science-fiction kind of future, but an actual forthcoming reality. AI is happening now!
Opinion is deeply divided on this issue. Some say: “AI is going to substitute human work; it is going to make us superfluous.” Others say: “But there will be more new jobs than substituted jobs.” From dystopia to utopia, every possible scenario is represented in this discourse. There are only two things for sure: Everyone is talking about AI and the future of work will indeed be significantly influenced by AI.
Therefore, it is time to take a look at the future of work and the change caused by AI. In order to provide a holistic view on this topic, this three-part blog series will do so through the lens of analytical philosophy.
How important work is for us is not obvious at first glance. We are used to grumbling about our stressful and annoying work. All too much we would like to give it up and devote ourselves to our free time and the important things in life. Many think about work in such a way and this mindset currently leads to a bigger discussion about the importance of work-life-balance and similar concepts. “So please AI, take over my work!! I don’t want it!” But work plays a central role in our lives. It is constituent for our identity and our social environment. In the society we live in, we are who we are because of our work.
These series of blogs will examine the question of whether AI can free us from work or will deprive us of work. They theorize on the future of jobs and ascertain the value of work in an AI-influenced world. This first blog will start off with a rather basic issue and prepares for the questions concerning the future of work:
- Why Philosophy of AI?
- Which jobs are going to be substituted by AI and will there be new jobs?
- How does a fundamental change in work impact our society?
- How is it possible to grant a fair distribution of wealth in our society in which some own all the productive capacity and others on nothing?
- How to be prepared for the change?
The aim of the philosophical approach is to explore the extensive ethical implications and social impact of AI. This will help to find a way to be prepared for the upcoming change of work and leads to the first and most fundamental question: Why Philosophy of AI?
Philosophy of AI: Why is this a good Approach?
While clearly distinct in approaches, technology (in general) and philosophy share the same object of interest: Humans.
The goal of technology development is to solve a specific practical problem faced in everyday life and thus to increase utility for humanity for the foreseeable future. But in most cases, the scope of technological development does not exceed the practical and current issue it is concerned with. There is just no need to do so if the problem in question can be technically solved. Technology always pursues one goal: Usefulness. It seems to be purely instrumental [1] and is rarely concerned about the side effects of its products.
Philosophy, in contrast, illuminates more than just current problems and practical aspects of human existence. In order to generate the broadest possible view on a certain topic, philosophical analysis considers not only the object of study itself, but also its ethical implications and other possible impacts on human affairs. Part of this is the examination of the emergence, the development and the nature of values. Therefore, scrutinizing and criticizing common positions and current developments, in order to locate changes in a distinct system of values is a major task in the field of philosophy.
Philosophy and Technology: A Contradiction or a Benefit?
In a nutshell: Philosophy usually introduces new questions and problems, whereas the purpose of technology, especially AI, is naturally to solve specific and already existing problems. Having this in mind a symbiosis between this two fields prima facie seems to be paradoxical.
But by posing more and new questions and critiquing proposed technological solutions and especially examining the underlying problem in a precise, philosophical manner, technology can provide long-lasting and more elaborate solutions. Philosophy provides the tools for this foresighted process, such as: logical analysis, ethical and moral expertise and a profound methodology of asking the right questions. In this case: How will AI impact the future of work?
This clearly compliments a prospective development of new technologies. When as many as possible implications of both the problem and the proposed technical solution itself are considered during its developmental process, future problems can be solved in a sustainable way. All of this applies to AI as a subset of technology, which for now should be defined as, “Science and engineering of making intelligent machines, especially intelligent computer programs,”[2].
But the relationship between AI and philosophy is much more far-reaching than that.
AI and Philosophy: A Special Relationship
The uniqueness of the relationship between AI and philosophy was already emphasized by computer scientist John McCarthy [3].While philosophy is complementary to all technical science in general, it is even foundational to AI as a special discipline and provides fundamental methodologies to this field.
Philosophers developed some of the most basic concepts of AI. Examples of this are, “… the research on features that an artifact must possess in order to judge it as intelligent,”[4] or the elementary concept of Rationality [4] that also originated from philosophical discourse.
More interesting in this context is the fact that philosophy is needed to guide the evolution of AI and organize its implementation in our lives, since more than trivial technologies, it implicates entirely new and unexplored ethical and social questions. Finding the right answers is the task of philosophy. But how?
From Micro to Macro: A Philosophical Way of Exploring
Analytical philosophy is about finding the most precise and at the same time holistic view on the explored topic. To achieve that, the following methodology will be applied: Beginning on a very atomistic level, this blog series will step by step detect all elements that are included by the question, “What is the impact of AI on the future of work?”.
Moreover, the relations, interdependencies, and intersections between them will be described, in order to recognize which factors influence what elements, in which way. From here the next step is to assemble all bits and pieces in a coherent way to create a model which represents the state of affairs.
Within this pattern it will be possible to alter variables or introduce new factors and observe how the different parts of the model are changing. This is were impacts and implications are made explicit. The observed change is then representing the answer this series is looking for: The impact of AI on the future of work. Based on this results we will also be able to make recommendation for policies concerning the future of work. Philosophy in this sense is an overall methodology that helps to understand the essence and because of that also the far-reaching impacts of a research object; of AI.
Our proposed step-by-step-analysis begins by illuminating what seems to be the driving factor in the context of this blog: What is that new technology that everybody is talking about?
Demystifying AI: A Pragmatic Definition
Now is the time to explore a fitting conceptual working framework of AI. What is artificial intelligence? Since there is not even a general definition of the term “intelligence”, it exceeds the scope of this blog to try to find an ideal definition of AI (even though it is certainly a very philosophical question). Instead it will be examined in which context AI is used and what it is supposed to do, giving at least a “pragmatic” definition of the term.
Fundamental to this technology is data (which by the way is also something that has implications like: privacy). It is the raw material which is needed to create artificial intelligence. There are basically two different approaches of using and processing this data in a meaningful way: Analytics and Machine Learning. The former (Analytics) helps to generate intelligence it allows to gain insights deriving from the underlying data. The other (Machine Learning) can make these insights actionable, which means it is becomes possible to create technological systems that are able to make decisions based on the given data.
AI is typically associated with Machine Learning and identifies patterns and schemes and make decisions that derive from this analysis. Based on probabilistic calculations machines learn to process information beyond what they were initially programmed or trained for. Unlike traditional computer-based solutions that rely on programming, AI does not rely on fixed rules and therefore can learn towards a goal in a meaningful way.
Even though the rapid pace of the development of AI technology should cause a broad debate, social and philosophical issues are still a blind spot in AI research [5]. In order to change this, the crucial philosophical question that is raised in this blog series is: How does AI affect the future of work?
The Link: Where AI affects Work
Terms like Cyber-Physical-Systems or Smart Factories are on everyone’s lips [6]. When talking about Industry 4.0 or the fourth industrial revolution, the prerequisite for this is always AI. Enterprises with large and often very complex systems are hard to manage. But since AI requires large amounts of data, which is certainly produced en masse in this context, an enterprise is the perfect environment for AI to develop its full potential. Hence an important area of application of AI will obviously be cost reduction and efficiency, especially its implementation as a process automation tool within enterprises.
Based on the relevant data, AI (i) can assist humans and facilitate processes, (ii) can augment humans/processes and (iii) can automate them. These are the roles that an AI can take on within a business process. In general, this can help to provide deterministic output for the enterprise. As the computer scientist Pedro Domingos points out: “Artificial intelligence is … the ability to solve hard problems.”[7] Problems that used to be solved by humans…
Within the context of an enterprise, AI is certainly an important driver of the change of the work. It first assists, then augments and finally automates working processes. This is a fact. We just don’t know to which extent and when this will happen.
To Be Continued…
Sources
- https://www.sfu.ca/~andrewf/books/What_is_Philosophy_of_Technology.pdf
- http://jmc.stanford.edu/artificial-intelligence/what-is-ai/index.html
- http://jmc.stanford.edu/articles/aiphil2/aiphil2.pdf
- V. Schiaffonati (2003): A Framework of the Foundation of the Philosophy of Artificial Intelligence; in: Minds and Machines, Vol.: 13.
- https://www.nature.com/news/there-is-a-blind-spot-in-ai-research-1.20805
- https://www2.deloitte.com/us/en/pages/technology-media-and-telecommunications/articles/artificial-intelligence-disruption.html
- https://www.scientificamerican.com/article/artificial-intelligence-will-serve-humans-not-enslave-them/
