New فارسی
Table of Contents
- Abstract
- Background
- Theoretical Framework Linking AI and Socialism
- Claims and Structure of the Argument
- Critiques and Counterarguments
- Conclusion
- References
Abstract
From a Marxist perspective, artificial intelligence, automation, and the transformation of the knowledge economy create new material and cognitive conditions for a possible transition from capitalist production toward democratic planning and more social forms of ownership. This capacity does not mean that socialism will arise automatically. The direction of change is determined by relations of ownership and power, the level of social consciousness, political freedom, and society’s ability to participate in decision-making. Advocates of this view argue that AI and its digital infrastructure turn knowledge, cognitive labor, and data into major productive forces and make coordination of production and social decision-making possible on a scale that was previously unattainable under capitalist appropriation [1][2][3].
In this framework, AI is not inherently emancipatory. Its social consequences depend on ownership, governance, and democratic control. Social or public ownership of AI platforms, data infrastructure, and strategic industries—combined with participatory planning and governance involving workers, knowledge producers, users, local communities, and citizens—could direct informational capacities toward human development rather than private profit [4][6]. Academic debate remains contested: without strong democratic institutions and collective rights, AI may reproduce or deepen asymmetries of power, surveillance, and control. The central question is therefore not whether AI by itself “produces socialism,” but under what political and institutional conditions its capabilities can support a freer, more equal, and democratically coordinated society [4][5][6].
Background
Marxist political economy assigns a central role to capital and the relations of production in shaping social and political life. Capital is not simply one resource among others; it is the dominant economic power in bourgeois society, organizing labor, investment, and the extraction of surplus value [8]. At each historical stage, transformations in production reveal both general features of human production and specific forms through which capital organizes work.
The passage from manufacture to the factory system altered the social division of labor and the relationship between agriculture and industry. Machinery expanded productive capacity but also intensified exploitation through longer or more tightly controlled working time and the commodification of labor power. Workers became increasingly subordinated to an objective system of production, while production itself became more concentrated, collective, and social [9]. The growth of factories and social labor required corresponding developments in communication and transportation, integrating production into wider markets and reorganizing the social metabolism of capitalist society.
For Marxist theory, the division of labor develops historically and gives rise to commodity relations and market exchange. The transition toward factory-centered organization is also a transition toward socialized production: products are made through vast networks of interdependence, even though their results are privately appropriated. This contradiction—social production alongside private control—opens the historical possibility that mature productive forces may be governed through more collective forms [10]. Scientific socialism does not treat that possibility as a prophecy. It analyzes how material capacities, class relations, institutions, and conscious political action interact [11][12].
Theoretical Framework Linking AI and Socialism
AI under Capitalism: Accumulation, Control, and Class Domination
AI is neither inherently liberating nor inherently oppressive. Yet within capitalist relations of ownership and power, it tends to serve accumulation, surveillance, and the consolidation of class domination. Patents, algorithms, software, platforms, cloud infrastructure, semiconductor supply chains, and data are controlled by a small number of firms. Technological power is therefore concentrated, while most people become dependent on digital systems produced under terms set by capital [13][15]. Rather than democratizing knowledge and power, AI can intensify monitoring, labor discipline, and the extraction of surplus value.
Digital Surplus Value and the Transformation of Labor
Value is increasingly extracted not only from formally waged work but also from users’ everyday digital activity. Clicks, searches, conversations, images, and interactions become data used for advertising, product development, and model optimization. Users often receive no direct wage for this contribution. Digital platforms thus extend value extraction beyond the traditional workplace into ordinary life, creating forms of invisible or unpaid labor [13]. AI also reorganizes paid work: it can deskill some tasks, intensify others, and move control over pace, evaluation, and scheduling from workers to opaque systems.
The Knowledge Economy and Monopoly Power
In the knowledge economy, information and intellectual assets become central to economic value. AI reinforces competition for technological, commercial, and military advantage, often directing innovation toward strategic control rather than inclusive development. Monopoly ownership of model weights, training data, computing capacity, and distribution platforms generates rents and barriers to entry. The result is a contradiction: knowledge is produced socially and cumulatively, but access to it is restricted through private property and technical enclosure.
Shaping Consciousness and Cultural Hegemony
AI systems do more than organize production. Search, recommendation, ranking, and generative systems help shape what people see, discuss, and regard as credible. Under concentrated ownership, these systems may reproduce dominant assumptions and narrow the field of political imagination. A democratic socialist approach must therefore address not only economic allocation but also pluralism, transparency, freedom of expression, access to knowledge, and society’s right to question and correct automated systems.
Toward Socialism: Democratizing AI and Redefining Value
The emancipatory possibility of AI begins with democratization: social control over essential infrastructure; public-interest access to data and computing; worker and citizen participation in design and deployment; and rules that distribute productivity gains through shorter working time, universal services, education, and social security. Such institutions would redefine value beyond private profitability and evaluate technology by its contribution to human flourishing, ecological sustainability, care, freedom, and collective capability [4][6][14].
Technology, Possibility, and Conscious Transition
Technological development creates possibilities, not predetermined outcomes. No productive force contains its own social destination. AI may strengthen capital or support democratic coordination depending on ownership, organization, law, culture, and collective action. A transition therefore requires social consciousness, durable democratic organization, and institutions capable of transforming ownership and governance. Treating technological advance as an automatic road to socialism would replace historical analysis with technological determinism.
Freedom, Pluralism, and Participation in Power
Any credible socialism for the AI age must place freedom and pluralism at its core. Democratic control cannot mean that a state bureaucracy or technical elite speaks on behalf of society. Workers, consumers, communities, researchers, minorities, and political currents must have real power to deliberate, contest decisions, obtain information, and change policy. Participation is not decorative consultation; it is a distribution of power, with enforceable rights, institutional checks, and the ability to revise technical systems.
Social Ownership Beyond State Ownership
Social ownership is broader than nationalization. It can include public institutions with democratic mandates, cooperatives, municipal and community ownership, commons-based infrastructures, public trusts, and mixed arrangements subject to transparent social goals. The decisive questions are who governs, who benefits, who can inspect the system, and whether society can correct its direction. State ownership without democratic accountability may reproduce hierarchy; private ownership with limited regulation may leave the central power of capital intact.
Claims and Structure of the Argument
Two broad socialist strategies are often proposed: first, socialization through nationalization of major industries and expansion of public ownership; second, radical decentralization and democratization of economic power. Either path can improve on capitalism only if it bypasses domination by private investment and places effective control in the hands of workers and citizens through workplace democracy, social control of investment, participatory budgeting, and accountable public institutions [4][6].
A recurring proposal is coordination through councils and federations rather than exclusive reliance on market price signals. In central planning, broad goals are established at the top and plans are revised through iterative information gathering. Participatory or democratic planning instead gives worker and consumer councils, intermediary facilitation bodies, and public deliberation a direct role in determining investment and output. Indicative prices or social costs can be adjusted through feedback loops to resolve mismatches between proposals, available resources, and needs [4][6].
Critics argue that central planning risks inefficiency and domination by a “coordinator class” of political and technical elites. Information flows upward while commands flow downward, potentially suffocating self-management. Participatory-economics proposals answer this objection through iterative negotiation among workers, consumers, and councils. Proposals are revised in light of feedback; investment choices are publicly debated; and final plans require democratic authorization. The claim is not that these processes are effortless, but that complex coordination can be organized without surrendering allocation to private capital or authority to an unaccountable bureaucracy [6].
Historical experience warns against identifying socialism with nationalization alone. Public ownership must be combined with democratic control, social supervision of investment, local empowerment, civil liberties, and mechanisms that prevent the reconcentration of power. Modern AI could reduce some informational and administrative burdens of planning, but it cannot decide social priorities. Models can compare scenarios; citizens must determine goals, rights, risks, and acceptable trade-offs.
Critiques and Counterarguments
Economic and Political Risks of AI Reproducing Capital
In Marxist analysis, AI can be understood as fixed capital: like machinery, it is introduced to reproduce and expand capital, not merely to facilitate production. Knowledge and automation may consequently be directed toward preserving dominant relations rather than egalitarian outcomes [2][7][13]. When controlled by a minority of owners, AI can intensify exploitation, surveillance, and concentration of power.
AI, Labor, and Accumulation
Automation may reduce direct labor inputs while preserving the claims of owners over output and income. Without democratic limits, the result can be unemployment, weakened bargaining power, intensified labor, and a larger separation between productivity and livelihood. The same technology could instead shorten working time and broaden access to social goods, but only if productivity gains are socially distributed [9][14].
Epistemological Risks
Modern science depends on codified and standardized protocols, but knowledge also rests on lived experience, practical judgment, and situated understanding. If knowledge production becomes fully mediated by algorithms, insights may be detached from their social contexts. “General intellect,” or collective knowledge, can be captured and redirected by capital, weakening the very social foundations required for democratic scientific development [11][13].
Limits of Technological Determinism
AI cannot be abstracted from its material and social environment. Systems optimized for profit and domination will tend to reinforce existing hierarchies, even when they appear technically neutral. Hardware production, energy consumption, mineral extraction, supply-chain labor, and environmental costs are also part of AI’s political economy [15]. An emancipatory program must therefore govern the entire infrastructure, not merely the behavior of consumer-facing models.
Unresolved Questions
More data and better modeling can support coordinated planning, but capacity is not legitimacy. Who sets objectives? How are minority rights protected? What information must remain private? How can citizens appeal an automated decision? How are ecological costs counted? How can local autonomy coexist with national and global coordination? Existing literature shows that technical progress becomes social progress only when underlying relations of capital, class, and political power are transformed [5][7][13].
Conclusion
AI and the information economy make knowledge, automation, and cognitive labor central productive forces. They can strengthen the material possibility of social ownership and rational democratic planning beyond traditional relations of capital [1][2][3]. But they are not an automatic agent of emancipation. Under concentrated capitalist ownership, AI is likely to deepen extraction, monopoly, surveillance, and unequal power.
A socialist horizon appropriate to the AI age therefore requires social or public control of critical infrastructure, cooperatives and commons, participatory management, public oversight, democratic planning, and robust political freedoms. AI may help societies process information, compare alternatives, and coordinate complex systems, but the transition itself depends on social consciousness, democratic organization, freedom and pluralism, participation in power, and a transformation of ownership and governance. Scientific socialism becomes more materially conceivable in the age of AI—not inevitable.
References
[1] Marx, Karl. Grundrisse: “Fragment on Machines” (1857–1858). Marxists Internet Archive.
[2] Marx, Karl. Capital, Volume I, Chapter 15: “Machinery and Modern Industry” (1867). Source.
[3] Engels, Friedrich. Socialism: Utopian and Scientific (1880). Source.
[4] Cockshott, W. Paul, and Allin Cottrell. Towards a New Socialism. Spokesman, 1993. Source.
[5] Davidson, Sinclair. “The Economic Institutions of Artificial Intelligence.” Journal of Institutional Economics 20 (2024). DOI.
[6] Albert, Michael, and Robin Hahnel. The Political Economy of Participatory Economics. Princeton University Press, 1991. JSTOR.
[7] Burns, Tony. “Marx, Automation and the Politics of Recognition within Social Institutions.” Critique 52, nos. 2–3 (2024): 357–378. DOI.
[8] Marx, Karl. Grundrisse: Introduction (1857). Source.
[9] Marx, Karl. Capital, Volume I, Chapter 14: “Division of Labour and Manufacture” (1867). Source.
[10] Marx, Karl, and Friedrich Engels. Manifesto of the Communist Party (1848). Source.
[11] Haug, Wolfgang Fritz. “General Intellect.” Historical-Critical Dictionary of Marxism, Rosa Luxemburg Stiftung, 2024. Source.
[12] Marx, Karl. Preface to A Contribution to the Critique of Political Economy (1859). Source.
[13] Fuchs, Christian. Digital Labour and Karl Marx. Routledge, 2014. DOI.
[14] Santoni de Sio, Filippo, Txai Almeida, and Jeroen van den Hoven. “The Future of Work: Freedom, Justice and Capital in the Age of Artificial Intelligence.” Critical Review of International Social and Political Philosophy 27, no. 5 (2024): 659–683. DOI.
[15] Valdivia, Ana. “The Supply Chain Capitalism of AI.” Information, Communication & Society (2025). DOI.
