Surveillance capitalism

Surveillance capitalism is the framing developed by Shoshana Zuboff — Harvard Business School professor emerita — that names the dominant business model of surveillance-based platform technology companies as a novel economic logic in which human experience is unilaterally appropriated as free raw material, translated into behavioral data, and traded as prediction products in what Zuboff calls behavioral futures markets. The framing's central mechanism is the concept of behavioral surplus: the byproduct of human activity online that exceeds what the platform needs to deliver its service, which it appropriates, processes through machine intelligence, and refashions into prediction products that anticipate what users will do next — products sold to advertisers and other buyers seeking to influence future behavior. The framing operates as an economic description of how surveillance platforms generate revenue, a political diagnosis of the power asymmetries that model creates, and a democratic-systems warning about the capacity for behavioral modification at scale that accumulating surplus eventually enables. It has become the primary English-language economic critique of the platform-capitalism business model in civil society advocacy across North America and Europe.

Origin

The framing first appeared in Zuboff's September 2014 essay "A Digital Declaration: Big Data as Surveillance Capitalism," published in German and English in the Frankfurter Allgemeine Zeitung, which characterised surveillance capitalism as "a radically disembedded and extractive variant of information capitalism" based on the commodification of human reality into behavioral data for analysis and commercial sale. The framing's peer-reviewed crystallisation came in her March 2015 paper "Big Other: Surveillance Capitalism and the Prospects of an Information Civilization" in the Journal of Information Technology (JIT 30: 75–89), winner of the 2016 International Conference on Information Systems Best Paper Award. The paper supplied the mechanism's precise anatomy: Google's AdWords system, recovering from the dot-com bust by monetizing behavioral exhaust from its search queries, had discovered that the actual raw material of its business was not the search service but the behavioral data users deposited as a byproduct. That byproduct — behavioral surplus — was fed into machine-intelligence manufacturing processes to produce prediction products whose commercial value derived from their accuracy about what users would do next. The products were sold in a new market species Zuboff named behavioral futures markets, completing the circuit from human experience to commercial revenue without users' knowledge or consent. "Big Other" named the resulting architecture of distributed power: a computational system that observes, interprets, and acts on behavior without being observable in return, producing a new form of domination through asymmetric knowledge rather than through coercion.

The framing's fullest elaboration is the 2019 book The Age of Surveillance Capitalism: The Fight for a Human Future at the New Frontier of Power (PublicAffairs; UK: Profile Books; 704 pp, ISBN 978-1-78125-685-5), published in the United Kingdom in October 2018 and in the United States in January 2019. The Financial Times called it "a masterwork of original thinking and research"; The New Yorker named it among its top nonfiction of 2019; Barack Obama listed it among his favorite books of that year. The book extended the mechanism into a systematic account: the graduation from behavioral prediction to behavioral modification ("actuarial" → "tuning" → "herding" products), Facebook's development of behavioral modification as surveillance capitalism's second stage after Google's discovery of the extraction logic, and the argument that the two companies' combined dominance over search, social media, mobile operating systems, video, and browser markets gave the surveillance-capitalism logic a structural grip on the primary channels through which contemporary public life is conducted. At its Harvard Gazette launch, Zuboff named the two channels through which the model threatens democracy: internally, behavioral modification at scale erodes the autonomy that moral judgment and democratic deliberation require; externally, the platforms' accumulating knowledge asymmetry over their users constitutes a new form of power without accountability to democratic oversight.

Civil society adoption

Amnesty International: "Surveillance Giants" (November 2019)

The framing's most consequential single civil society adoption in the 2019–2020 window was Amnesty International's report "Surveillance Giants: How the Business Model of Google and Facebook Threatens Human Rights" (Index POL 30/1404/2019, November 2019). Amnesty — whose primary mandate runs to political imprisonment, torture, and the death penalty rather than platform data practices — entering the surveillance-capitalism argument was itself significant: it marked the framing's translation from an economic-and-democratic-theory register into the international human rights law register, with the attendant machinery of rights obligations, state duties, and corporate accountability under international law. The report declared that the companies' "surveillance-based business model is inherently incompatible with the right to privacy" and poses "an unprecedented danger to human rights," mapping the extraction mechanism across the Universal Declaration of Human Rights framework — five specific rights: privacy, freedom of opinion and expression, freedom of thought, equality and non-discrimination, and control over personal information — and demanding "a radical transformation of the tech giants' core business model" alongside government enforcement of data protection law that prohibits conditioning service access on consent to data surveillance for marketing. The Cambridge Analytica scandal and documented electoral manipulation were named as downstream effects of the behavioral modification capacity the model accumulates. The Amnesty report became the framing's most-cited civil society deployment in the subsequent EU regulatory campaign period.

European digital-rights campaigning

In Europe, the framing moved through AlgorithmWatch, the European Digital Rights network (EDRi), and Access Now as part of the coordinated civil society campaign to ensure the EU AI Act embedded binding rights prohibitions rather than voluntary AI-ethics compliance mechanisms. The surveillance-capitalism critique supplied the economic-systems layer the campaign needed: the ethics-washing framing named why voluntary commitments were structurally insufficient; the surveillance-capitalism framing named the underlying economic incentive structure that made voluntary commitments economically self-undermining for companies whose revenue model required behavioral surplus extraction. The two framings operated as a diagnostic pair across the AI Act legislative campaign through 2021–2023. Privacy International deployed the framing in its campaign on the adequacy of data protection law for the surveillance-capitalism era, focusing on the gap between the GDPR's consent-based framework and a business model in which meaningful individual consent is structurally impossible: the population-level aggregate of behavioral data is the product, and no individual data-subject opt-out can address a system in which the opt-out decision is itself a behavioral signal in the prediction apparatus.

Why it has carried

Three features explain the framing's reach across civil society, academic, and regulatory registers.

First, the framing supplies an economic-logic critique that operates at the level of the system's architecture rather than its individual failure modes. The cascade of surveillance-capitalism-era controversies — Cambridge Analytica, the Facebook election-interference hearings, the YouTube radicalisation research, the Instagram teen-mental-health studies — each constitute individually documentable harms, but the framing supplies the structural logic explaining why they are not anomalies but properties of a system optimised for behavioral modification at scale. The civil society organizations that adopted the framing did so because it enables a diagnosis that survives individual company responses intact: fixes that address surface manifestations leave the underlying behavioral-surplus extraction logic — and its incentive to expand the scope and depth of behavioral observation — untouched.

Second, the framing's Harvard institutional provenance gave it unusually rapid uptake across policy, civil society, and academic registers simultaneously. Most conceptual innovations in the AI accountability field arrive from activist or academic positions that face institutional resistance before achieving policy uptake. The surveillance-capitalism framing arrived from a senior Harvard Business School professor, already in the Financial Times and on Barack Obama's reading list, with a full 700-page scholarly apparatus, which compressed the trajectory from critique to policy consideration and let civil society organizations cite it as authoritative in regulatory submissions without first building the citation base independently.

Third, the framing's 2019 timing was calibrated to a pre-formed public attention window. The book's US publication landed in the aftermath of the Cambridge Analytica scandal (March 2018), the Facebook Congressional hearings (April 2018), and the GDPR's first enforcement actions (beginning May 2018) — a window in which the structural architecture of platform data practices had become a live public policy problem and civil society organizations were actively building the vocabulary for the EU AI Act and Digital Services Act regulatory campaigns that ran through 2021–2023.

Limits and critiques

The framing has attracted two substantive criticisms within the digital rights and AI accountability field.

First, critics have argued that the surveillance-capitalism framing's focus on the advertising-based economic model of US platform companies understates the state-surveillance side of the behavioral-data extraction problem. This critique holds that "capitalism" in the concept's name points analysis toward market and antitrust solutions — break up Big Tech, regulate behavioral advertising — while bracketing the more structurally intractable fact that the same data infrastructure serves immigration enforcement, predictive policing, and national-security surveillance. The global-south critique runs a parallel argument: the framing's primary empirical cases are Google and Facebook in North American and European markets, which leaves the behavioral-extraction logic running in regulatory environments with fewer democratic-accountability levers unaddressed.

Second, the adjacent data-colonialism framing — developed in Latin American civil society organising over the same 2019–2022 period — explicitly positions itself as a supplement to and partial critique of the surveillance-capitalism frame. Where surveillance capitalism names the economic logic of data extraction from the perspective of the platform's behavioral-futures market, data colonialism names the political-economic asymmetry between the jurisdictions in which platforms are headquartered and the jurisdictions in which their disproportionate impacts land. The two framings converge in naming extraction as the operative mechanism but diverge on which extraction asymmetry is the analytical priority: economics-of-attention, or decolonial political economy.