“Algorithmic violence” refers to a new form of power constructed by contemporary digital capitalism through the appropriation of data, platform monopolies, and algorithmic governance. Unlike the Hobbesian Leviathan, which exercises power through visible coercion, or the Orwellian model of overt surveillance and repression, algorithmic violence operates through a softer and more sophisticated form of domination.
Capital functions as the hidden sovereign, algorithms as its instrument of power, and digital platforms as the institutional infrastructure through which this power is exercised. Individuals appear to possess more choices than ever before, yet increasingly lose the capacity to determine what they see, what they believe, and why they believe it. The greatest danger of algorithmic violence therefore lies not in its explicit destruction of freedom, but in its ability to disguise control as convenience, discipline as personalization, and obedience as individual choice. Its ultimate achievement is the production of a social environment in which people experience themselves as free precisely while their autonomy is being progressively diminished.
Digitalization has not dissolved the fundamental power relations of capitalism. Instead, it has provided capital with new technological mechanisms for accumulation and domination. In digital capitalism, data functions as a new form of productive resource, platforms constitute new centers of economic and social power, and algorithms provide the crucial mechanism through which economic power is transformed into the capacity to shape social behavior.
Capital continuously captures the traces individuals leave behind through searches, clicks, browsing histories, purchases, interactions, locations, and even emotional responses. These traces are transformed into calculable and commercially exploitable assets. Smart devices, social media, and the Internet of Things have extended this process from public spaces into the most intimate dimensions of everyday life. Individuals are increasingly rendered visible, measurable, predictable, and governable.
The problem is therefore not merely the loss of privacy. It is the emergence of a social condition in which individuals know that their behavior may constantly be observed, recorded, analyzed, and evaluated. Such awareness can produce a chilling effect: people modify what they say, what they search for, and what they do because they anticipate possible consequences. Power no longer needs to prohibit an action directly. It only needs to create the perception that one is being watched. Surveillance thus becomes self-discipline, and external control becomes internalized restraint.
Data accumulation subsequently becomes platform monopoly. Companies such as Google, Meta, Amazon, Apple, and Microsoft have consolidated enormous economic and informational power through network effects, economies of scale, proprietary infrastructures, and control over user data. Traditional monopolies primarily involved control over physical means of production and markets. Digital monopolies extend this power to the infrastructures through which information, social relations, and public attention are organized. The more users a platform attracts, the more data it collects; the more data it collects, the more precisely its algorithms can predict and personalize; the more personalized its services become, the more users it attracts. This self-reinforcing cycle produces a form of digital enclosure in which a small number of private corporations increasingly determine the conditions under which social communication takes place.
Algorithms transform this economic concentration into a deeper form of social power. Data can record behavior, but algorithms can predict, classify, rank, and influence it. Algorithms are not politically or economically neutral. Their objectives, training data, optimization criteria, and recommendation systems embody particular institutional priorities. When platforms optimize for clicks, engagement, watch time, and advertising revenue, algorithms are structurally encouraged to privilege whatever captures attention rather than whatever is most accurate, rational, or socially valuable. Technological “optimization” thus becomes a mechanism of capitalist discipline.
This discipline begins at the level of cognitive input. Individuals are confronted with an unprecedented abundance of information, yet abundance does not necessarily produce knowledge. Personalized recommendation systems increasingly replace active exploration with automated selection. Algorithms infer users’ preferences from previous behavior and continuously provide similar content, thereby producing filter bubbles and information cocoons.
Confirmation bias is consequently reinforced by technological design: individuals encounter more information that confirms what they already believe and fewer arguments that challenge their assumptions. Society appears to inhabit one interconnected information space, while in reality citizens increasingly occupy different algorithmically constructed realities.
The consequences become even more serious when algorithms contribute to the production of simulated reality. Under an attention-driven business model, factual accuracy is not necessarily what determines visibility. Content capable of generating outrage, fear, excitement, or controversy often performs better than complex and carefully contextualized information. Fake news, conspiracy narratives, political manipulation, and synthetic media can therefore acquire substantial visibility.
The political controversies surrounding the 2016 U.S. presidential election and the British referendum, as well as the subsequent proliferation of AI-generated political deepfakes, demonstrate that algorithms do not merely filter reality; they can increasingly participate in producing alternative versions of reality. When citizens can no longer reliably distinguish between fact, opinion, propaganda, and synthetic fabrication, the shared epistemic foundation necessary for democratic politics begins to erode.
Algorithmic domination does not stop at determining what people see. It increasingly influences how they think. In the attention economy, complex arguments are structurally disadvantaged because they require time, concentration, and cognitive effort, whereas short, emotionally charged, and highly stimulating content can be consumed and circulated almost instantaneously. Deep reading and sustained reflection are gradually displaced by fragmented consumption.
Algorithms do not need to prevent people from thinking. They only need to make thinking appear unnecessary. When individuals become accustomed to receiving preselected information, simplified explanations, and algorithmically generated conclusions, independent judgment is gradually replaced by cognitive outsourcing.
The result is not conventional ideological indoctrination but a subtler form of cognitive impoverishment. Individuals retain the formal right to think for themselves while increasingly losing the intellectual conditions necessary to do so. Critical thinking requires exposure to disagreement, uncertainty, complexity, and competing interpretations. Yet algorithmically organized environments tend to reward familiarity, immediacy, and emotional confirmation.
Users are therefore encouraged to remain within cognitive communities whose members share similar assumptions. These communities provide belonging and emotional security, but they also intensify polarization and distrust toward outsiders. Social differences are transformed into epistemic divisions, while political disagreement becomes a conflict between mutually exclusive identities.
The transformation of cognition becomes explicitly political when algorithms begin to govern emotional attention. Digital platforms systematically reward high-arousal emotions such as anger, fear, resentment, and moral outrage because such emotions generate interaction and prolong engagement. As a consequence, public discourse can undergo a process of adverse selection in which information with the greatest social value is displaced by information with the greatest capacity for circulation. Complex policy questions receive less attention because they lack the dramatic simplicity required by the attention economy, while polarizing narratives and symbolic conflicts are continuously amplified.
Political identity consequently becomes a form of digital currency. Divisions between left and right, red and blue, insiders and outsiders, are easier to circulate than nuanced arguments about public policy. Individuals are encouraged to interpret political questions through the lens of group belonging rather than independent judgment. The algorithmic organization of attention thus transforms social difference into commercial value: polarization produces engagement, engagement produces data, and data produces profit.
At this point, algorithmic violence begins to threaten the foundations of democracy itself. Democracy requires more than elections and formal political rights. It presupposes citizens capable of forming relatively autonomous judgments, accessing diverse information, expressing disagreement, deliberating with others, and translating their preferences into collective decisions.
If individuals’ informational environments, political emotions, and cognitive frameworks are systematically shaped by privately controlled algorithms, formal political freedom may increasingly coexist with substantive political dependence.
The first consequence is the weakening of democratic decision-making. Traditional democratic theory assumes that citizens express political preferences and that institutions aggregate these preferences through procedures of representation and deliberation. Algorithmic systems introduce a disturbing reversal: citizens may increasingly be represented not by what they consciously express, but by what platforms infer about them.
Political preferences can be predicted from behavioral data, segmented into micro-targeted audiences, and acted upon before individuals fully recognize those preferences themselves. The citizen risks becoming less a political subject than an object of political computation.
This creates the possibility of what might be called algorithmic representation.
Platforms can simulate public opinion by aggregating behavioral traces, engagement statistics, and emotional reactions. Yet the resulting “public voice” may not represent autonomous political judgment at all. It may simply reflect the outcomes of algorithmic selection and capitalist optimization. The danger is that the appearance of democratic participation conceals the displacement of genuine participation. Citizens are given the sensation of being represented while their capacity for collective self-government is gradually weakened.
The second consequence concerns freedom of expression. Digital platforms have undoubtedly lowered the barriers to political speech. Yet the right to speak and the ability to be heard are no longer equivalent. Platform governance determines not only what users may post, but also how widely their speech can circulate. Content moderation, recommendation systems, ranking mechanisms, account restrictions, and differential visibility create a hierarchy of attention. A citizen may possess the formal right to speak while lacking any meaningful ability to reach a public audience.
This transforms the classical conception of the marketplace of ideas. Digital platforms are not neutral spaces in which competing opinions automatically receive equal opportunities to be heard. They are privately governed communication infrastructures whose rules are often opaque and whose economic incentives influence the distribution of visibility.
Expression can therefore become commodified: attention itself becomes a scarce resource, and those with greater economic or technological capacity can acquire greater amplification. Freedom of speech risks being transformed from an equal political right into a competition for algorithmic visibility.
The third consequence is the erosion of political efficacy. Digital participation often creates the appearance of empowerment. Citizens can like, share, comment, sign petitions, join online campaigns, and participate in virtual debates within seconds. Yet symbolic participation does not necessarily translate into institutional influence. Individuals may experience an immediate sense of political agency while remaining largely disconnected from the processes through which public policy is actually made. This produces a paradox of digital democracy: the easier political participation becomes, the less capable participation may become of changing political reality.
When repeated digital participation fails to generate substantive institutional responses, political frustration can gradually become political apathy. Citizens may begin to doubt whether their actions matter, whether institutions are responsive, and ultimately whether democratic participation itself has meaningful value. The most effective form of domination is therefore not necessarily the prohibition of resistance, but the creation of a political environment in which resistance appears futile.
Algorithmic violence ultimately exposes a deeper crisis concerning the meaning of freedom. Classical freedom involves both freedom from arbitrary interference and the positive capacity to form judgments and act according to one’s own will. Under digital capitalism, however, freedom risks becoming a preconfigured experience.
Individuals are not necessarily deprived of choices; rather, the choices presented to them are increasingly structured by systems they cannot see or control. They are not necessarily forced to adopt particular opinions; rather, the informational environment in which opinions emerge is increasingly engineered in advance. They are not prevented from speaking; rather, the conditions determining whether their speech becomes socially consequential are controlled by private algorithms.
Yet this critique should not become a form of technological determinism. Algorithms themselves are neither inherently emancipatory nor inherently oppressive. The decisive question is the power structure within which technology operates. The same digital technologies that can facilitate knowledge, communication, and democratic participation can also deepen monopoly, surveillance, manipulation, and polarization. The problem is therefore not technology as such, but the concentration of technological power in the hands of private capital without adequate democratic accountability.
Escaping algorithmic violence consequently does not require abandoning technology. It requires breaking the exclusive alliance between capital and algorithmic power and restoring human agency over the infrastructures that increasingly organize social life. Algorithmic systems must become subject to meaningful transparency, data rights, institutional accountability, public oversight, and democratic regulation. Most importantly, citizens must regain the capacity to understand, question, and contest the mechanisms through which algorithms shape their informational and political environments.
Socrates used questioning to force individuals to confront assumptions they had never examined; contemporary algorithms use recommendation to make those examinations appear unnecessary. The contrast captures the fundamental tension between classical freedom and digital power. Freedom does not mean merely possessing access to unlimited information. It means retaining the capacity to decide what deserves attention, to encounter what challenges one’s convictions, to reflect before responding, and to determine one’s political judgment without invisible manipulation. The central struggle of the algorithmic age is therefore not between humans and machines, but between human autonomy and the concentration of technological power.
If algorithms remain instruments of capital accumulation, they may transform democratic freedom into a carefully managed illusion. But if technological power is brought under democratic control, algorithms can instead become instruments for expanding human knowledge and political participation. The essential task is thus to ensure that algorithms remain tools of human beings rather than human beings becoming objects of algorithms. Only then can digital civilization fulfill its emancipatory promise instead of becoming a new and subtler architecture of domination.
Source: ssaj ajcass, cass, oxford acadmic, epale, scalevise
