The 21st Century: A 2024 Update

8 min readOct 17, 2024

“But the overall pattern is clear: in case after case, when a model can be created and tested, it tends to perform as well as, or better than, human experts making similar decisions. Too often, we continue to rely on human judgment when machines can do better.” — Andrew McAfee, Erik Brynjolfsson[1]

As life expectancy increases, one of the unexpected benefits is that even monumental events repeat. For the second time, I am going through an industrial revolution. Coincident changes in technology, economics and social constructs mark each industrial revolution. In the 1960s I saw firsthand the emergence of the Third Industrial Revolution (3IR). The 3IR was marked by:

· the commercialization of computers and computerization,

· the large scale development of multinational business models, and

· the emergence of widespread activism for minority and women’s rights

We are now in the early stages of the Fourth Industrial Revolution (4IR). The current 4IR was announced by the World Economic Forum in 2016 but probably began with research success at Stanford as early 2010. This research transitioned artificial intelligence from largely a university research topic to the commercially viable technology that launched the 4IR. I believe AI (and machine learning) will be comparable to mathematics, a tool of unimagined creative utility to discover new knowledge. Not only will AI create new science(s) and better engineering, but today the most effective business and operating models use AI. The social manifestations in the 4IR are still unclear, but self-employment, personal healthcare and privacy might emerge as the most important issues.

The social change brought about by a new industrial revolution is always the hardest factor to anticipate and predict. This time understanding social change is made harder by several complex factors. The polarization we all sense is typical for the beginning of each industrial revolution, but exacerbated today by our network-centric lifestyle, bad actors, the unstable economic situation in China and extremist splinter groups. Additional pressures always arise from the inherent inequality of every economic system (the Matthew effect).

Further complicating our understanding of the current situation is the transition in our theory of knowledge (epistemology) from materials-based to information-centric. John Archibald Wheeler’s seminal 1990 paper, “Information, Physics, Quantum: The Search for Links”[2], is an important contribution to this new perspective on reality. This paper is colloquially identified as the “It from Bit” paper. Physical reality (‘it”) is derived from information (‘bit”). Wheeler postulates that physical reality is derived from information, a binary choice based on measurement and observation. This approach reinforced the quantifiable nature of reality at the particle level. Information at this particle level expanded the range of scientific research across a wide range of sciences with the emergence of more advanced AI models.

As math, physics and information science advanced, or “merged”, several important fundamentals became apparent:

· Science was transformed from equations to computational models with stochastic computation replacing deterministic equations.

· The sciences moved from the study of equilibriums to a focus on non-equilibrium phenomena, reflecting the nonlinear, chaotic nature of many systems. (Weather models, protein folding and economic modeling are examples.)

· Network-centric, collectivist organizations such as cities emerged as the principal drivers of advanced economic and social behavior, reinforcing holistic systems thinking as a necessary paradigm.

· All of this complexity put an increased emphasis on transdisciplinary approaches to problems and the use of multi-modal data in AI models.

This transition to new epistemology, new science and new computational models contributes to the increasing complexity and consequent uncertainty shaping society today. David Wolpert and Kyle Harper at Santa Fe Institute argue that advances in gathering energy and processing information inevitably lead to a more complex society. “So the more computational power society has, the more energy extraction per capita, and vice versa.”[3] These inputs advance sequentially in their own step functions similar to industrial revolutions. A simple, straight forward explanation of the increased complexity we see today.

Today we have unheard of levels of information processing. According to Statista, data creation will have increased 90 fold in 15 years.[4]

2010–2 zettabytes

2020–64.2 zettabytes

2025 forecast — 181 zettabytes

Computation used to train large AI models has increased from hundreds to billions of petaFlops, an increase of 10⁵.

These trends put increased pressure on the need for renewable energy and seriously challenge the Paris Agreement 2015 for a temperature increase limited to 1.5°C. In 2010, renewable energy provided 425 billion kilowatt hours (kWh) of electricity, which was 10% of total U.S. electricity generation.[5] By 2023, renewable sources accounted for about 21% of total utility-scale electricity generation.[6] The trend in renewable energy is encouraging but the growth does not match the computation statistics especially when we look at the large data centers planned by leading corporates like OpenAI and Microsoft. Available renewable energy needs to increase. Long overdue is a national strategy.

We all appreciate that we live in an increasingly network-centric society. We have been warned that ChatGPT and similar algorithms will increase corporate efficiency and productivity but will lead to increased job losses as positions disappear. Probably alarmist, but the consulting firm McKinsey estimates that fifty (50) percent of all jobs will be lost to computer automation.[7] This scale will only add to the social unrest, increase the demands for guaranteed or universal basic income and move the U.S. more toward a socialist system perhaps with democracy still in place. News on job creation is much less publicized, but the CHIPS Act investments foster job creation and could create new “industrial clusters” with collaborative potential for additional job creation. Eventually, the private sector will accelerate their investment in applied AI business models, which will help to offset job loss elsewhere.

The transformative changes in healthcare are underestimated in considering job loss. Regardless of the advances in robotics, I believe medicine will remain a “hands-on” business. AI is moving medicine from diagnostics and treatments to a preventive science with resultant expectations for increased life expectancy. These senior citizens will need increased attention in increasing numbers. The number of Americans ages 65 and older is expected to increase from 58 million in 2022 to 82 million by 2050, representing a 47% increase.[8] Medicare spending alone is expected to reach $8.3T[9] by 2040 and reach 5.5% of U.S. GDP by 2053.[10] Healthcare will be a positive factor for job creation and potentially the largest industry job creator. The U.S. Bureau of Labor Statistics forecasts healthcare to contribute 40 percent of new jobs in the period 2022 to 2032.[11] I encourage my university students to focus on healthcare and the related sciences and computer science, computation and engineering as attractive 21st-century career opportunities. Environmental science and related fields are also encouraged, especially as environmental degradation contributes to health issues.

Job opportunities raise the question of what education will need to provide. With an increased emphasis on science, computation, math, computer science and AI, we should start teaching science and math much earlier, at a higher standard. The first two years of these subjects could be offered more widely in high schools and the existing curriculum could be offered earlier to junior high school students. Worried that this approach is not consistent with declining performance on national and international tests, maybe break down education in junior high school into three tracks:

1-Science, Math and Computer Science

2-College Prep

3- Trades and Services

Category-1 develops the new science and engineering to keep us alive or even improve life. Category-2 manages everything and Category-3 provides all the support for daily life. Everybody would take courses at least in introductory AI.

The biggest assumption in this education scenario is whether students will still “attend” school (buildings). If robots solve the babysitting issue, schools as physical space may be unnecessary. Regardless of the physical outcome, I hope that schools transition to more online self-directed learning, more active learning and more mentor/coaching time. Today my university students ask for more and more active learning, internship opportunities and mentoring. Much of the mentoring and coaching in the future could be done by robots. I also think that gaming for education will increase dramatically. Grandview Research forecasts that the educational games market will grow at a CAGR of 38.6 percent from 2024–2031 and reach $21.6B in 2031. Other analysts would call the forecast conservative. Educational games could support my expectation for much more self-directed learning.

We have not updated our thinking on education since Horace Mann’s (1843) original work. We need to transition from computer “training” to the computer as a “partner” or a resource for self-directed learning. Part of this learning will need to focus on Augmented Collective Intelligence (ACI), MIT professor Thomas Malone’s term for the emergent results of real time network connectivity between humans, AI, robots, automation and cybersecurity. Whereas AI might be described as mimicking human behavior, ACI will enhance human capabilities and decision making. Education will need to prepare students for this change in the work environment, value creation and customer experience. This is particularly important as an increasing number of people are self-employed, by choice or for other reasons. Forbes estimated that by 2030, 30 percent of individuals in the U.S. will be self-employed. This forecast was reached eleven years ahead of the estimate.[12] Challenging times are ahead for the workforce, educators, politicians and forecasters. Iterative approaches are recommended.

The purpose of this article is not to predict the future but rather to give the reader a set of assumptions that should guide their thinking and behavior in the 21st century.

“We live in a time where all current systems of human thought from physics to politics are dominated by a set of underlying assumptions that are mostly outdated and no longer work in a very different world that is more interdependent and connected than ever before in human history.” — Fritjof Capra[13]

BUCKLE UP!!

Previous writings on related topics include:

1. https://roberthhacker.medium.com/a-primer-for-the-21st-century-the-5-systemic-changes-7e710e3c2321

2. https://roberthhacker.medium.com/escaping-descartes-ai-complexity-and-a-new-scientific-paradigm-5538bc17805a

3. https://roberthhacker.medium.com/the-necessity-of-government-crafting-a-clear-strategy-for-the-future-936e029fc1f7

References

[1] Machine, Platform, Crowd: Harnessing Our Digital Future …

[2] https://philpapers.org/archive/WHEIPQ.pdf

[3] https://fasterplease.substack.com/p/the-computational-power-of-complex?publication_id=232077&utm_campaign=email-post-title&r=qp6q&utm_medium=email

[4] https://www.statista.com/statistics/871513/worldwide-data-created/

[5] https://www.c2es.org/content/renewable-energy/

[6] https://www.eia.gov/energyexplained/electricity/electricity-in-the-us-generation-capacity-and-sales.php