AI and THE DEATH OF WORK 1:
This Is Not a Repeat of Schumpeter’s Cycle of “Creative Destruction” and Economic Rebirth
David Barnhizer
Nikolai Kondratiev, Joseph Schumpeter, and Jack Ma
The CEO of China’s technological giant Alibaba, Jack Ma, warned in 2017 that given the rise of AI/Robotics in reducing available work opportunities, we are in for multiple decades of economic and social turmoil that may settle down and smooth out by mid-century. Ma stated:
“Social conflicts in the next three decades will have an impact on all sorts of industries and walks of life. [adding] “A key social conflict will be the rise of artificial intelligence and longer life expectancy, which will lead to an aging workforce fighting for fewer jobs.”
What is occurring with AI/Robotics is not only unique but outside Austrian economist and Harvard University professor Joseph Schumpeter’s process of “Creative Destruction”, a cycle of innovation that Schumpeter once stated he wished he had called “transformation” rather than “destruction”. In the classic Schumpeterian model, pre-existing economic forms are replaced in large part by new technologies, and then after a difficult period better forms of work eventually become the new norm of work and production, at least for some people. See, Joseph Schumpeter, Capitalism, Socialism and Democracy.
Nikolai Kondratiev anticipated Schumpeter when he produced an analysis for Joseph Stalin who had directed him to prove the inevitable collapse of Capitalism and the superiority of Marxism. Unfortunately for Kondratiev he ended up describing the fact that Capitalist economies undergo periodic large-scale technological declines and resurgences on a roughly fifty-year basis that Joseph Schumpeter later named “Kondratiev Waves” in honor of Kondratiev’s insightful analysis.
The problem was that rather than a permanent collapse, Kondratiev’s analysis concluded there is an eventual productive rebirth mechanism within Capitalism represented by new and improved forms of production and work stimulated by the innovative dynamics created amidst the bottom part of the Long Wave cycle. Kondratiev’s findings and influence as a highly regarded agricultural economist caused him to be arrested and imprisoned in 1930 under what are considered false charges by the Soviet government of being part of a non-existent political group. After years in prison he was executed in 1938 in Stalin’s Great Purge.
Schumpeter’s cycle of economic transformation and re-creation is neither neat nor painless for those who lose out as their world disintegrates. There is significant turmoil and displacement for many people as the world of work shifts on its axis. Those displaced by the transformation suffer, but in a macro-economic sense the overall community ends up benefiting as new types of jobs are created in large numbers.
In past transformations, this process of creative destruction has balanced out over time so that labor and skills were able to move generally from obsolete activities to new positions that added value to the recreated economy. When the automobile moved from a limited technological oddity to widespread acceptance as a personal conveyance, blacksmiths and wagon makers were highly paid skilled trades. They became more and more irrelevant as the industrial revolution of the early 20th century moved forward. But many of the same skills developed by blacksmiths also provided value for the displaced individual by satisfying new demands and exploiting new opportunities made possible by the emergent technologies.
While prior industrial revolutions displaced workers, the pace of change was such that many blacksmiths could transfer their skills to the machinist and fabricator trades, and wagon makers could at least in some cases transform their production facilities from producing wagons to producing cars. Even today the changes tend to be matters of scale and category rather than complete replacement. Blacksmiths and farriers continue to provide services to horse owners and some have gone into metal art and ornamental décor. The danger is not simply in absolute and complete elimination of work functions. The threat is the dramatic reduction in the number of jobs and the types of skills required to function productively in the transformed economy.
The “AI/Robotics” Revolution
During the computer-based industrial revolution that began in the 1960s and early 1970s, we have seen a different picture than that of the industrial revolution of the late-1800s and early-1900s. While employment for real wages for low and middle income workers remained largely flat since the late 1960s, productivity per worker increased significantly from the advent of significant introduction of computers, computer-based automation, and robotics in the mid-1970s. These developments and a comprehensive analysis of AI and its implications are considered at length in a book I co-authored with my son Daniel Barnhizer, Professor of Law at Michigan State University. See, The Artificial Intelligence Contagion: Can Democracy Withstand the Imminent Transformation of Work, Wealth and the Social Order? (Clarity Press, 2019).
With AI/robotics we are experiencing a massive and accelerated disruption in the market for human labor that takes place over an extremely compressed time scale compared to all previous so-called “revolutions”. The ongoing and accelerating AI/robotics-based industrial revolution will render many more jobs worthless even while increasing productivity with ever fewer human inputs. This means that data on labor productivity levels are inevitably deceptive in terms of measures of employment. This is because it is a classic “doing more with less” situation in which far fewer workers are needed to produce what is required and per worker data does not reflect what happens to workers displaced by individual workers who utilize the new technology.
Many still believe, or choose to believe for a variety of reasons, that the creative destruction process will follow past patterns. I continually read reports and position papers that assert that after a time, the system will create new jobs for those put out of work while augmenting the ability of others to do the jobs they already have. In the context of the effects of AI/Robotics which is a technological form we have never before encountered as already being revealed by ChatGPT and DALL-E advances (with much more to come)—this is an intellectual trap.
The impacts being experienced are categorical in terms of the types of jobs most affected, and territorial because some cities, regions, and nations will be dynamic centers of development while others will be left in the lurch. As the AI/Robotics “Fourth Industrial Revolution” unfolds, there will be hotspots of economic activity such as Silicon Valley, Seattle, the North Carolina Research Triangle, the “Beltway Bandits” of the DC area, Boston, and some other dynamic clusters. One report indicated the most dynamic and best performing areas of economic growth included Provo Utah, Raleigh North Carolina, and Dallas. Big gainers were the Palm Bay-Melbourne area in Florida, Olympia Washington and Hickory North Carolina. Paul Davidson, “Provo’s economy ranks best in U.S.”, USA Today, 1/10/18.
The fact is that many areas of the US will be left behind and their residents will have little, if any, growth opportunity or chance at social mobility. Regardless of the geographic location—economically vital or not--far fewer jobs will exist in absolute terms, and there will be a limited number of high quality jobs in any location. Numerous areas of the country will become economic “dead zones”. Given the demographics, dwindling resources and inequalities, some cities will become unsustainable and dangerous.
Another aspect of what is occurring is that, although we might like to think that our jobs will only be replaced when it is established that automated, computerized or robotic systems can do a qualitatively “better” job, it is neither inevitable nor necessary that the new systems that replace us must perform “better” than humans, or even as well. While quality of performance is a consideration to a degree, we delude ourselves if we think it is the only element or even the primary one in many instances.
We should, for example, ask ourselves whether the endless hoops we have to jump through to connect with a human when seeking information by calling a business or government agency is “better” than that of the human receptionists replaced by the computerized system. The AI-based systems simply need to be less expensive, more reliable, not miss work because of sick time and vacation absences, not require maternity leave, and offer employers much less hassle and expense than is presented by human workers and unions.
The ChatGPT Phenomenon Is Just Beginning
“Artificial Intelligence Could Automate Two-Thirds of All American Occupations: Goldman Sachs,” Katabella Roberts, April 18, 2023. https://www.theepochtimes.com/artificial-intelligence-could-automate-two-thirds-of-all-american-occupations-goldman-sachs_5202670.html. “[E]conomists Joseph Briggs and Devesh Kodnani said that a new wave of AI systems, such as ChatGPT, could have a “major impact” on employment markets across the globe, while advances in such technology could trigger shifts in workflows that could “expose the equivalent of 300 million full-time jobs to automation.”
“AI could grow so powerful it replaces experienced professionals within 10 years, Sam Altman warns: Both OpenAI's Sam Altman and Google's Sundar Pichai published essays in support of regulating AI”, Emma Colton, 5/25/23. https://www.foxnews.com/tech/ai-grow-powerful-replaces-experienced-professionals-10-years-sam-altman-warns. “Altman, the chief of the AI lab behind popular platforms such as ChatGPT, … [warned] that "we must mitigate the risks of today’s AI technology. "It’s conceivable that within the next ten years, AI systems will exceed expert skill level in most domains. …”
“WEF survey says world will shed 14 million jobs by 2027”, Aislinn Murphy, 5/1/23. https://www.foxbusiness.com/economy/wef-survey-world-shed-14-million-jobs-2027. “Over 80% of surveyed companies reported being "likely" or "highly likely" to incorporate digital platforms, education and workforce development technologies or big data analytics by 2027. … The survey involved over 800 companies across various industries around the world.”
“ChatGPT AI lists jobs it can do better than humans as millions could be put out of work: ChatGPT was unveiled in November and went on to break records as the fastest-growing user base”, Emma Colton, 4/5/23. https://www.foxbusiness.com/technology/chatgpt-ai-lists-jobs-it-can-do-better-humans-millions-could-be-put-out-work.
"OpenAI’s wildly popular chatbot, ChatGPT, is expected to replace 4.8 million U.S. jobs, according to a new report. Outplacement and executive coaching firm Challenger, Gray & Christmas recently asked ChatGPT’s bot a series of questions, including "What jobs can ChatGPT replace?" and what fields the bot would be most capable of working. … [When asked] The bot told the outplacement firm that it would most likely replace positions that are repetitive and predictable, and ones that are also steeped in language requirements. Those fields, according to the bot, include: customer service representatives; translators and interpreters; technical writers; copywriters; data entry clerks. Challenger, Gray & Christmas crunched the number of jobs in each listed field using data from the Bureau of Labor Statistics and found that at least 4.8 million American jobs could be replaced. The AI chatbot added that it could see itself entering other fields such as data science; machine learning; mathematics and statistics; computer science; robotics and automation; and business.”
The ChatGPT technology was even asked what type of jobs it could replace. Emma Colton explains an answer. “When asked what jobs could be replaced by AI broadly, ChatGPT said financial analysts and customer service representatives, among others, could potentially be replaced by AI in the future. Some examples specifically given by ChatGPT include transportation, manufacturing and financial services. Other industries that could be at risk include healthcare in which "AI can assist with diagnosing and treating patients, reducing the need for some healthcare workers" and journalism where "AI can generate news articles and summaries, potentially impacting jobs."ChatGPT AI lists jobs it can do better than humans as millions could be put out of work”, Emma Colton, 4/5/23.
“AI Will Massively Affect White Collar Jobs: Study”, Petr Svab, 3/28/23. https://www.theepochtimes.com/ai-will-massively-affect-white-collar-jobs-study_5154399.html?ea_src=frontpage_A_topstory&ea_med=list.
“Many white-collar jobs can be done much more efficiently with the help of artificial intelligence such as ChatGPT, according to a recent research paper. … the impact would tend to be greater on higher-paid jobs that require higher education, the paper says. … The most affected professions would be mathematicians, tax preparers, financial quantitative analysts, writers and authors, web and digital interface designers, interpreters and translators, survey researchers, public relations specialists, and animal scientists. The most affected industries would be various market and financial services, real estate and insurance services, data analysis, media, and occupations revolving around writing, math, and coding, according to the paper.”
How Bad Will the Job Loss Be? Predictions of Accelerating Job Loss
We need to gain a clear and honest understanding of what we face. Taken together with the job loss trends mentioned above, the predictions offered below represent what a range of experts have said concerning what is expected to occur over the next few decades by way of an AI/robotics “revolution”. If the predictions are even fifty percent accurate, we face problems with which we are unlikely to be able to deal effectively. The experts’ predictions include the following projections.
50% of US jobs will disappear by 2030. [Fast Forward 2030: The Future of Work and the Workplace] or 50% of US jobs will be gone by 2025. [2 Billion Jobs to Disappear (Globally) by 2030].
5,000,000 US manufacturing jobs have already been lost since 2000.
12,000,000 US jobs will disappear by 2026. [Robots Set to Disrupt White-Collar Work, 2016].
500 million jobs in the world’s 25 richest economies will be lost to AI/robotics.
50 percent of today’s work activities will be automated by 2050—give or take 20 years. Robots will take over most of the world’s jobs by 2045.
4.1 million US driver compensated jobs (taxis, semi-trucks, delivery vehicles, busses) will be lost to self-driving vehicles.
Imposing a $15 minimum hourly wage in the US could eliminate between three to five million jobs. [Earned Income Tax Credit Is Better Tool to Raise Income than $15 Minimum Wage, 2016].
Black males in the US have an unemployment rate of 17.5% and Black teenagers an unemployment rate of 41%. Youth unemployment is also threatening the future of millions of young Europeans.
The unemployment rate in the Eurozone reached 10% in 2009, “and has been stuck in double digits ever since. On average, more than one out of five young people in the labour force are unemployed, but in the worst hit crisis countries, almost half of people looking for work can’t find jobs.” (From Nobel Laureate in Economics, Joseph Stiglitz).
The McKinsey Global Institute has warned that if the “slow growth” conditions of the past decade continue, up to 80 percent of people in developed economies could see flat or falling incomes. MIT researchers report a large-scale “hollowing out” of the US middle class with many sliding down on the socio-economic ladder. While working hours are at an all-time high human productivity has collapsed and earnings are flat or declining for most workers even as the cost of goods and services continues to increase.
Predictions by the World Economic Forum, Goldman Sachs, the World Bank
The World Economic Forum (WEF) estimates 14 million net job losses worldwide by 2027: Artificial intelligence and economic and social conditions are expected to result in 83 million job losses, with only 69 million jobs created over the next five years. Julien Bouissou, May 3, 2023. https://www.lemonde.fr/en/economy/article/2023/05/03/world-economic-forum-estimates-14-million-net-job-losses-worldwide-by-2027_6025227_19.html#.
Like every innovation, artificial intelligence (AI) fuels fears of job destruction. While the ChatGPT program is arousing the interest of the general public and awakening the appetite of investors, a World Economic Forum survey published on Sunday, April 30, estimated that 14 million jobs will be lost worldwide on balance over the next five years (69 million created and 83 million destroyed), i.e. 2% of total employment. It anticipates that nearly a quarter of employees will change jobs, mainly due to the advent of new technologies. The authors of the study based their estimates on information collected from 803 companies worldwide employing 11.3 million people. The most significant job creation is in the education sector (+10%), and agriculture (especially machine operators), while job losses are mainly in administrative tasks, such as accounting services.
“Goldman Sachs Predicts 300 Million Jobs Will Be Lost Or Degraded By Artificial Intelligence”: Jack Kelly, Senior Contributor, March 31, 2023. https://www.forbes.com/sites/jackkelly/2023/03/31/goldman-sachs-predicts-300-million-jobs-will-be-lost-or-degraded-by-artificial-intelligence/
If generative AI lives up to its hype, the workforce in the United States and Europe will be upended, Goldman Sachs reported this week in a sobering and alarming report about AI's ascendance. The investment bank estimates 300 million jobs could be lost or diminished by this fast-growing technology. Goldman contends automation creates innovation, which leads to new types of jobs. For companies, there will be cost savings thanks to AI. They can deploy their resources toward building and growing businesses, ultimately increasing annual global GDP by 7%. In recent months, the world has witnessed the ascendency of OpenAI software ChatGPT and DALL-E. ChatGPT surpassed one million users in its first five days of launching, the fastest that any company has ever reached this benchmark.
Goldman predicts that the growth in AI will mirror the trajectory of past computer and tech products. Just as the world went from giant mainframe computers to modern-day technology, there will be a similar fast-paced growth of AI reshaping the world. AI can pass the attorney bar exam, score brilliantly on the SATs and produce unique artwork. While the startup ecosystem has stalled due to adverse economic changes, investments in global AI projects have boomed. From 2021 to now, investments in AI totaled nearly $94 billion, according to Stanford’s AI Index Report. … Office administrative support, legal, architecture and engineering, business and financial operations, management, sales, healthcare and art and design are some sectors that will be impacted by automation.
Foreword: World Development Report 2019 (World Bank)
Adjusting to the changing nature of work also requires rethinking the social contract. We need new ways to invest in people and to protect them, regardless of their employment status. Yet four out of five people in developing countries have never known what it means to live with social protection. With 2 billion people already working in the informal sectorunprotected by stable wage employment, social safety nets, or the benefits of educationnew working patterns are adding to a dilemma that predates the latest innovations.
This Report challenges governments to take better care of their citizens and calls for a universal, guaranteed minimum level of social protection. It can be done with the right reforms, such as ending unhelpful subsidies; improving labor market regulations; and, globally, overhauling taxation policies. Investing in human capital is not just a concern for ministers of health and education; it should also be a top priority for heads of state and ministers of finance. The Human Capital Project will put the evidence squarely in front of those decision makers, and the index will make it hard to ignore.