This reflection by Mahmoud Rasmi is based on the monthly Tech Leaders Salon discussion, where we discuss books on topics including AI, philosophy, strategy, management, and leadership. If you’re interested in joining the conversation, you can learn more about and join the discussion group here.
From Past Disruptions to the AI Era
What does a world without work look like, and how ought we to deal with it at a policy and governance level? These are some of the questions Daniel Susskind's A World Without Work addresses. Susskind surveys the impact of technological innovation and task automation on human employment, drawing an important distinction between past technological disruptions and current AI-driven ones.
Technological advancements are usually met with mixed reactions, ranging from extreme advocates and early adopters to an opposing pole of skeptics and opponents. Underlying this spectrum are justifiable arguments about how technology improves our lives and drives economic growth on the one hand, and concerns over how the changing landscape could lead to job loss, impacting the livelihoods of many people by rendering their craft or profession obsolete or redundant on the other.
Susskind argues that in the past, the social and existential anxieties triggered by automation proved to be unfounded because machines augmented human labor, increased production at scale, and almost always created new jobs. His main argument is that this pattern doesn't quite hold in the AI era because technology will slowly take on not only routine tasks, but also non-routine tasks, complex activities, and cognitive work.
Due to the paradigmatic shift in AI development, we seem to have managed to develop machines and systems that mimic human reasoning, relying on big data to carry out tasks like driving, medical diagnosis, legal analysis, writing, mathematical and data analysis, software development, and more.
The more complex and sophisticated this technology becomes, which seems to work without relying on human-like consciousness, the more task encroachment spans manual, cognitive, and affective domains, leading to human displacement. As a result, there exists a possibility that with AI, more people will find themselves without work.
Automation in the age of labor always led to more growth and an increase in the size of the "economic pie," with these gains being redistributed in society as more jobs were created, new skills were in demand, wages increased in certain sectors, and so on.
Susskind argues that although AI will continue to expand the size of the economic pie, solving the problem of production, it will break the link between economic growth and job creation. It will slowly replace humans without creating jobs at the same rate as before. As a result, capital will be concentrated among the few AI labs that own the means of production. A minority would also partake in this economic pie because they have the relevant skills required to keep things working, but the great majority would find themselves without work due to the absence of demand.
Society will be confronted with three major problems: inequality of wealth distribution, the concentration of power, and a lack of purpose and meaning due to job loss. Susskind writes:
Technological unemployment, in a strange way, will be a symptom of that success. In the twenty-first century, technological progress will solve one problem, the question of how to make the pie large enough for everyone to live on. But, as we have seen, it will replace it with three others: the problems of inequality, power, and purpose.
Technological progress brings prosperity, but AI will cause three problems that we have to deal with. Susskind dedicates the third part of the book to examining these problems and suggesting potential solutions that require coordinated action from policymakers, governmental and private institutions, and society as a whole.
Education, Wealth, and the Big State
In the short term, the best way to tackle these problems is by rethinking our educational system, changing what and how we teach to adapt it to the demands of AI. This includes offering an interdisciplinary curriculum aiming to cultivate both technical and soft skills to equip people with the abilities needed to thrive in routine and non-routine jobs. The concept and aim of education ought to change too, from merely a stage we go through early in our lives to acquire knowledge and skills to a lifelong learning process where we continuously alternate between periods of learning and work to adapt to changing market demands and the latest technology.
Revamping education merely offers a short-term antidote. In the long run, when work grows scarce and technological unemployment turns more structural, inherent to the new AI world, it yields a society without work. In such a world, Susskind argues that societies must aim to construct a Big State that shares asset ownership with private corporations (some sort of citizens' wealth fund) and taxes high earners and corporate profits, redistributing the income via a Conditional Universal Basic Income scheme. The scheme would require citizens to engage in community and meaningful work, contributing to the overall well-being of society, especially since paid labor would have largely receded.