The discussion surrounding Artificial Intelligence (AI) often oscillates between utopian techno-optimism and apocalyptic dread, largely confining the debate to the realms of computer science and immediate economic disruption. Yet, to grasp the true magnitude of the AI revolution, we must look beyond the algorithms and server farms; we must turn to historical sociology. When a technology fundamentally rewrites the means of production, it inevitably rewrites the social contract. To understand our current global trajectory, we cannot rely on predictive coding. We must dust off the analytical tools used to dissect the last great shift: the Industrial Revolution.
By applying the methodology of Barrington Moore Jr.’s seminal work, “Social Origins of Dictatorship and Democracy”, to the digital age, a clear picture emerges. We are not merely facing a shift in how we work; we are standing at a critical juncture that will dictate the political regimes of the 21st century. The digital transformation of the economic base is actively laying the groundwork for the political superstructure of tomorrow, forcing us to ask whether this new era will serve as the great democratizer or the ultimate engine of authoritarian control.
AI as a general-purpose technology and the industrial benchmark
AI has moved to the center of global debate, but viewing it merely as a technological innovation misses the main point. It represents a general-purpose technology—much like the invention of electricity or the steam engine—that acts as a transformational force cutting horizontally across every social, economic, and even security sector.[1] The scale of this shift is massive. Projections from consulting firms such as McKinsey[2] and PwC[3] suggest that by 2030, AI could inject US$15.7 trillion into the global economy, boosting global GDP by 14%. These figures underline a reality that goes beyond the obvious: AI is not just tweaking existing systems; it is fundamentally disrupting and reconstructing them. We see this everywhere, from sharpened diagnostic accuracy in healthcare[4] and personalized curricula in education[5] to a labor market being overhauled by automation and layoffs.
This AI-driven paradigm shift bears a striking historical similarity to the 19th-century Industrial Revolution. The First Industrial Revolution, by shifting production from human and animal muscle to machine power (steam), triggered the first period in history where per capita income increased steadily, independent of population growth. While global trade volume was negligible in the early 1800s, industrialization-fueled mass production pushed global exports to roughly 8-9% of world output by 1913.[6] As a result, supply and demand balances extended beyond national borders, transforming into a truly global market system. This significant expansion of economic infrastructure naturally led to changes in the political and social structures. Before the Industrial Revolution, production was mainly agrarian and spread out. The rise of factory systems caused people to move to cities. In England, urbanization increased dramatically from 20% to 70% during the 19th century.[7]
Structural breaks from steam power to cognitive automation
This demographic density, paired with a shift in capital ownership, wrestled the nature of politics away from the exclusive grip of the aristocratic elite. As the bourgeoisie rose as a new economic power and the working class swelled in numbers, both groups carved out space in the political arena to amplify their demands. This dynamic laid the groundwork for the rise of mass political parties and the expansion of suffrage, as exemplified by the Reform Act of 1832.[8] Ultimately, the process ignited by the steam engine redesigned not just factory floors but also parliaments and social contracts. Today, AI is poised to act as a similar catalyst. It has the potential to fundamentally alter not just algorithms but also the very mechanics of diplomacy, culture, and governance.
While debates continue over whether AI could trigger a transformation as profound as the 19th-century precedent, the data points unmistakably to a structural break. Even though the news warns of a potential AI bubble akin to the dot-com crash, the often visualized impact through on-the-ground realities is undeniable.[9] Goldman Sachs reports that Generative AI could automate nearly 300 million full-time jobs globally.[10] Meanwhile, the IMF forecasts that 60% of employment in advanced economies faces exposure to AI-driven risks or transformation.[11] These numbers signal something far deeper than mere sectoral efficiency gains; they imply a fundamental rethinking of labor markets and, by extension, social stratification.
The obsolescence of the middle class and the digital Leviathan
So, how will this seismic shift in the economic infrastructure reshape the political superstructure? We find a compelling framework in historical sociology, specifically in Barrington Moore Jr.’s master work, “Social Origins of Dictatorship and Democracy.”[12] Moore argued that during the transition from agrarian to industrial societies, the elimination of the peasantry and the specific alliances forged between landed elites and the bourgeoisie determined whether nations drifted toward democracy or dictatorship. His famous maxim, “no bourgeois, no democracy”, warrants reconsideration in the age of AI.
Today, AI is driving a process analogous to Moore’s elimination of the peasantry, but this time, it is the obsolescence of the white-collar middle class. If cognitive automation renders the middle class economically redundant, the collapse of democracy’s main pillar could shove global politics down a far more authoritarian path. At this point, the state’s reflex aligns historically with the principle of raison d’état. The core motivation is simple: ensure system continuity and stave off chaos. However, AI grants states unprecedented surveillance capabilities. China’s Social Credit System stands as a stark illustration of how this technology can morph into a digital Leviathan in the hands of the state.
Bifurcation of the future: the democratic vs. authoritarian path
We face Moore’s historical bifurcation once again: The democratic path, in which AI democratizes efficiency, empowers civil society, and enhances transparency; the authoritarian path: the state weaponizes AI as a tool for suppression and absolute control, justifying it as a defense against systemic threats like disinformation, cyberattacks, and civil unrest. Currently, the state’s anxiety over governability is tipping the scales toward the second path—a centralized model where security trumps liberty. While AI’s automation of routine and cognitive tasks is hailed as a technological triumph, the flip side reveals a looming threat of deep structural unemployment. The World Economic Forum’s (WEF) 2023 report projects the displacement of 83 million jobs against the creation of only 69 million over the next five years. This net loss signals an unequivocal contraction and a severe skills mismatch in the labor market.[13]
The liquidation of routine roles bears a striking resemblance to the enclosure movement in England, as analyzed by Moore. In the 18th century, landless peasants flocking to cities formed industrial capitalism’s reserve army of labor and a hotbed for potential rebellion. Today, masses rendered obsolete by AI are accumulating as the new dispossessed of the digital age. The most vulnerable link in this chain is the developing world, with its youth bulge. Data from the International Labor Organization (ILO) shows that global youth unemployment is already three times that of adults. By eliminating entry-level roles—the very rungs young people use to climb into the workforce—AI could only exacerbate this divide. One of Moore’s central theses in “Social Origins of Dictatorship and Democracy” is that the failure to integrate the peasantry during modernization (as seen in China and Russia) inevitably leads to peasant revolts and revolutionary violence.
Digital feudalism and the new “iron and rye” alliance
Today, particularly in the Global South—nations like India, Türkiye, and Brazil—the plight of the educated yet unemployed youth mirrors the sociological profile of Moore’s peasantry: a mass demanding radical change. Moore posited that when an old order (the labor-intensive economy) collapses without a new, inclusive order (the AI-driven economy) to replace it, the resulting vacuum is inevitably filled by violence and radicalism. Consequently, the exclusion of young people from the workforce is not merely an economic statistic; viewed through a Moorean lens, it is a political powder keg. If states fail to channel the surplus value generated by AI into funding or retraining these redundant masses, historical precedents suggest this demographic could become either the foot soldiers of far-right populism or the agents of chaotic and unpredictable social explosion.
Moore’s analysis originally focused on how the organization of agrarian production—land ownership, serfdom, tenancy—dictated political regimes. When we transpose this perspective to the 21st century, we see that data has replaced land, and algorithmic production has supplanted agricultural production. The political economy of the digital age bears a startling resemblance to the feudal or semi-feudal structures Moore dissected. Today, the giants of Big Tech are the new lords of these digital domains. Meanwhile, billions of users act as modern serfs, generating content or simply leaving digital footprints for the harvest, yet possess no property rights over the value they create.
The digital feudalism[14] hypothesis poses a critical question through the lens of Moore’s theory of class alliances: How will the new ruling class—the tech elite—engage with the state? Moore observed that in England, a compromise—however contentious—between the landed aristocracy and the rising bourgeoisie gave rise to democracy. In Germany, conversely, the iron and rye alliance between these classes forged an authoritarian path—a revolution from above. We face a similar risk of revolution from above in the age of AI. Tech giants and the state apparatus could align on a foundation of data monopolies[15] and security imperatives, steering us toward what Moore termed the fascist path. In this scenario, the quid pro quo is clear: the state grants tech company regulatory immunity and market protection; in return, these companies provide the capacity to surveil citizens and manipulate opposition through algorithmic censorship. This is a digital iteration of the “capitalism without democracy” model that Moore identified in Germany and Japan—a modernization that consolidates the bond between state and capital while systematically excluding civil society.
Charles Tilly, another titan of historical sociology, famously linked modern state formation to war-making.[16] He argued that, to wage war, states had to extract taxes and control populations, a necessity that gave rise to modern bureaucracy. The AI revolution elevates Tilly’s concept of state-making to a new dimension. Today, state power is measured less by artillery and ballistics, and more by computing power and data processing capacity. While the traditional state was charged with guarding physical borders, the AI-empowered state could guard—or manipulate—cognitive borders. This fundamentally alters the social contract. In Moore’s era, the state extracted taxes and conscripts in exchange for theoretical security. In the AI era, the state extracts data in exchange for predictability. But this predictability comes at the steep price of individual liberty.
Tools like predictive policing and algorithms designed to preempt social unrest are transforming the state’s monopoly on violence into something unprecedentedly preventive and all-encompassing.[17] The brute force used to suppress Moore’s peasant revolts is being replaced by an invisible, omnipresent digital discipline. This is Michel Foucault’s panopticon, upgraded for the silicon age. The state is evolving from an entity that crushes rebellion into one that eliminates the very probability of rebellion. By rendering the friction and bargaining essential to democracy impossible, this dynamic fosters a governance model that is frictionless, efficient, and fundamentally unfree.
The agony of delayed modernization in the Global South
Barrington Moore placed special emphasis on the agony of late modernization. In countries like Russia and China, the pressure exerted on the peasantry by rapid, state-enforced industrialization sparked revolutions. Today, nations undergoing a delayed AI revolution face a parallel peril. While North America and Western Europe (and arguably China) are the producers of AI technology, the vast majority of the Global South serves merely as consumers—or worse, as data colonies.[18] Seen through Moore’s lens, modernization in these regions is not organic; it is a technological standard imposed from the outside. This stifles the development of a native bourgeoisie (a local tech class), transforming domestic elites into comprador structures heavily dependent on global tech giants.
More critically, when the youth unemployment crisis in developing nations collides with AI automation, it could trigger exactly the kind of plebeian and anti-capitalist backlash Moore warned about. If countries like India, Brazil, or Türkiye lose their competitive edge in labor-intensive sectors—textiles, call centers, assembly lines—the resulting economic shock could overwhelm existing political regimes.[19] In these contexts, AI will not be seen as a tool for efficiency, but as an engine of mass impoverishment. The petty bourgeois radicalism and status anxiety that Moore identified as the roots of fascism could easily fuel modern neo-Luddite movements or drive a retreat into hyper-nationalist or religious restorationism.[20] When writing “Social Origins of Dictatorship and Democracy”, Moore’s core concern was simple: Who picks up the tab for the transition to modernity? In democracies, the cost was amortized over time and relatively softened; in dictatorships (communist or fascist), the bill was paid via the violent liquidation of the peasantry.
The AI revolution presents humanity with a similar invoice. Who pays this time? The obsolete middle class? The data-colonized developing world? Or individuals stripped of their privacy? Moore’s historical materialism does not offer a deterministic blueprint; rather, it underscores the critical importance of choices and alliances. If the 19th-century bourgeoisie and proletariat fought the aristocracy for democratic rights, the 21st century demands a similar battle line. This time, the fight is not for land ownership, but for data ownership, algorithmic transparency, and cognitive sovereignty. If states follow Moore’s authoritarian path—viewing AI solely as a tool for regime security and forging secret digital iron-and-rye alliances with Big Tech—the political character of the 21st century could be digital totalitarianism.
However, there is an alternative. If civil society, unions, and the new cognitive proletariat can coalesce to demand democratic oversight, we might forge a new social contract—perhaps through mechanisms like Universal Basic Income and Property—that broadens the distribution of AI-generated wealth. Ultimately, AI is not destiny; it is a policy battleground. As Moore noted, all radical changes start with violence, but the direction and outcome of that violence are determined by the political choices of social classes and the coalitions they build. Humanity currently faces the greatest leverage point since the invention of the steam engine. Whether this leverage elevates democracy or fortifies top-down, modernizationist dictatorships could be the defining political science question of the coming years.
Conclusion
In the final analysis, the shape of the AI-driven future is determined more by societal dynamics than by the technology itself. Moore showed us that the path to modern democracy was marked by intense, often violent, struggles over class interests, just as the path to fascism was paved with elite cooperation and systematic suppression of the masses. Today, the players have shifted. Big Tech replaces the landed aristocracy, data replaces arable land, and the displaced cognitive worker takes the place of the uprooted peasant, but the underlying structural forces remain the same. If we succumb to technological determinism, allowing a revolution from above where state security apparatuses and monopolistic tech lords dictate the terms of transition, the digital age will culminate in a frictionless but absolute totalitarianism.
However, recognizing this historical parallel is our greatest safeguard. The responsibility now lies with a new democratic coalition: a digitally conscious civil society, organized labor, and the emerging cognitive workforce, all demanding cognitive sovereignty and fair redistribution of algorithmic benefits. The steam engine built the modern world, but human political choices decided who was crushed beneath its wheels and who directed its course. As we stand on the edge of this new era, the most important question is no longer what AI can do, but what political realities we will permit it to shape.
[1] Timothy Bresnahan, “Artificial Intelligence Technologies and Aggregate Growth Prospects,” in Prospects for Economic Growth in the United States, 1st ed., ed. John W. Diamond and George R. Zodrow (Cambridge University Press, 2021), https://doi.org/10.1017/9781108856089.008.
[2] Jacques Bughin et al., “Notes from the AI Frontier: Modeling the Impact of AI on the World Economy,” Discussion Paper (McKinsey Global Institute, 2018), https://www.mckinsey.com/featured-insights/artificial-intelligence/notes-from-the-ai-frontier-modeling-the-impact-of-ai-on-the-world-economy.
[3] Anand S. Rao and Gerard Verweij, “Sizing the Prize: What’s the Real Value of AI for Your Business and How Can You Capitalise?,” Global Artificial Intelligence Study (PwC, 2017), https://www.pwc.com/gx/en/issues/analytics/assets/pwc-ai-analysis-sizing-the-prize-report.pdf.
[4] “Can AI Improve Medical Diagnostic Accuracy?,” HAI Stanford, October 28, 2024, accessed January 27, 2026, https://hai.stanford.edu/news/can-ai-improve-medical-diagnostic-accuracy.
[5] Morgan Kelly, “AI Can Deliver Personalized Learning at Scale, Study Shows,” November 2025, https://home.dartmouth.edu/news/2025/11/ai-can-deliver-personalized-learning-scale-study-shows.
[6] United Nations Conference on Trade and Development, Trade and Development Report, 1997: Globalization, Distribution and Growth, Report UNCTAD/TDR/17 (United Nations, 1997), https://unctad.org/system/files/official-document/tdr1997p2_en.pdf.
[7] Urbanization – Industrial Revolution, Population, Infrastructure, Britannica, accessed January 28, 2026, https://www.britannica.com/topic/urbanization/Impact-of-the-Industrial-Revolution.
[8] Matthew A. McIntosh, The Victorian Middle Class and the Industrial Revolution, Economics, October 29, 2025, https://brewminate.com/the-rise-of-the-victorian-middle-class-industrialism-and-the-shaping-of-the-modern-era/.
[9] David Streitfeld, “The A.I. Boom Is Unlike the Dot-Com Boom,” December 9, 2025, https://www.nytimes.com/2025/12/09/technology/ai-boom-unlike-dot-com-boom.html.
[10] Joseph Briggs and Devesh Kodnani, “The Potentially Large Effects of Artificial Intelligence on Economic Growth,” Goldman Sachs, 2023, https://www.goldmansachs.com/insights/articles/generative-ai-could-raise-global-gdp-by-7-percent.
[11] Mauro Cazzaniga et al., “Gen-AI: Artificial Intelligence and the Future of Work,” nos. 979-8-40026-248–2, International Monetary Fund, 2024, https://www.imf.org/en/Publications/Staff-Discussion-Notes/Issues/2024/01/14/Gen-AI-Artificial-Intelligence-and-the-Future-of-Work-542379.
[12] Barrington Moore, Social Origins of Dictatorship and Democracy: Lord and Peasant in the Making of the Modern World, with a new foreword by Edward Friedman and James C. Scott (Beacon Press, 2015).
[13] The Future of Jobs Report 2023, World Economic Forum, 2023, https://www.weforum.org/publications/the-future-of-jobs-report-2023/.
[14] UCL Institute for Innovation and Public Purpose, “Digital Feudalism: The Future of Data Capitalism, YouTube, 2021, 01:33:20, https://www.youtube.com/watch?v=fKgPQSa1_0o.
[15] Jonathan Joseph, “We Need to Talk About Data Monopolies,” TechPolicy.Press, November 27, 2024, accessed January 28, 2026, https://www.techpolicy.press/we-need-to-talk-about-data-monopolies/.
[16] Charles Tilly, “War Making and State Making as Organized Crime,” in Bringing the State Back In, ed. Peter B. Evans et al. (Cambridge University Press, 1985), Cambridge Core, https://doi.org/10.1017/CBO9780511628283.008.
[17] Maria Lungu, “Predictive Policing AI Is on the Rise – Making It Accountable to the Public Could Curb Its Harmful Effects,” The Conversation, May 6, 2025, https://theconversation.com/predictive-policing-ai-is-on-the-rise-making-it-accountable-to-the-public-could-curb-its-harmful-effects-254185.
[18] Nick Couldry and Ulises Ali Mejias, The Costs of Connection: How Data Is Colonizing Human Life and Appropriating It for Capitalism, Culture and Economic Life (Stanford University Press, 2019).
[19] Carl Benedikt Frey and Michael A. Osborne, “The Future of Employment: How Susceptible Are Jobs to Computerisation?,” Technological Forecasting and Social Change 114 (January 2017): 254–80, https://doi.org/10.1016/j.techfore.2016.08.019.
[20] Shoshana Zuboff, The Age of Surveillance Capitalism: The Fight for a Human Future at the New Frontier of Power, First edition (PublicAffairs, 2019).
