What passes for AI accountability often props up the industry: many critiques and reforms act as decoys that conceal the networks of power and extraction making AI possible; to secure real fairness we must reckon with the material political economy behind AI, not just its surface practices.

Reckoning with the Political Economy of AI: Avoiding Decoys in Pursuit of Accountability

Janet Vertesi, danah boyd, Alex Taylor, Benjamin Shestakofsky · April 17, 2026

arxiv commentary n/a evidence 7/10 relevance Source PDF

The paper argues that apparent critiques of AI often function as industry-friendly 'decoys' that obscure the material political economy and networks of power underpinning AI development, and that meaningful accountability requires confronting those political-economic structures directly.

The Project of AI is a world-building endeavor, wherein those who fund and develop AI systems both operate through and seek to sustain networks of power and wealth. As they expand their access to resources and configure our sociotechnical conditions, they benefit from the ways in which a suite of decoys animate scholars, critics, policymakers, journalists, and the public into co-constructing industry-empowering AI futures. Regardless of who constructs or nurtures them, these decoys often create the illusion of accountability while both masking the emerging political economies that the Project of AI has set into motion, and also contributing to the network-making power that is at the heart of the Project's extraction and exploitation. Drawing on literature at the intersection of communication, science and technology studies, and economic sociology, we examine how the Project of AI is constructed. We then explore five decoys that seemingly critique - but in actuality co-constitute - AI's emergent power relations and material political economy. We argue that advancing meaningful fairness or accountability in AI requires: 1) recognizing when and how decoys serve as a distraction, and 2) grappling directly with the material political economy of the Project of AI. Doing so will enable us to attend to the networks of power that make 'AI' possible, spurring new visions for how to realize a more just technologically entangled world.

Summary

Main Finding

The paper argues that meaningful accountability for AI requires shifting scrutiny away from isolated technical fixes and towards the material political economy that makes “AI” possible. What appears to be critique—technical audits, definitional debates, safety research, calls for limited regulation—often functions as a set of decoys that divert attention from the network-making power of financiers, firms, infrastructure owners, and other elites. Those decoys are integral to the Project of AI: they stabilize markets, enroll critics, and enable accumulation and consolidation of wealth and power.

Key Points

Project of AI: The authors frame “AI” as a world-building project in which firms, financiers, infrastructure providers, and states assemble new networks (compute, chips, data centers, talent, capital) that instantiate durable power relations.
Network-making power: Drawing on Castells, Actor–Network Theory, and economic sociology, the paper highlights how heterogeneous actors configure social, material, and financial networks to shape markets and governance.
Markets are both found and made: The current AI market emerged amid uncertainty (post-2008, pandemic, ChatGPT moment) and is actively being constructed by actors who shape rules, funding flows, and infrastructure to entrench advantage.
Decoys: The authors identify five decoys that misdirect attention from political economy while helping to co-constitute it:
- Ontological decoy — debates about “what AI is” maintain semantic ambiguity that helps actors claim relevance and capture funding.
- Inevitability decoy — narratives that AI’s progress or impacts are inevitable reduce appetite for structural interventions.
- Disruption decoy — framing change as entrepreneurial disruption legitimizes displacement and concentration without addressing redistribution or power.
- Safety decoy — focusing on technical safety/alignment as the primary intervention deflects from structural reforms (ownership, labor, energy).
- Regulatory decoy — industry-led or narrow regulatory proposals (technical standards, self-regulation) can appear to be accountability while preserving incumbent advantage.
Decoys are not incidental; they enlist researchers, journalists, funders, and regulators into stabilizing the Project of AI, creating accountability traps.
Two-step corrective: (1) recognize when critiques function as decoys, and (2) grapple directly with the material political economy—i.e., who controls compute, capital, data, labor, and infrastructure.

Data & Methods

Approach: Conceptual and theoretical synthesis rather than new quantitative empirical analysis.
Sources and traditions: Interdisciplinary literature review drawing from communication studies (Castells), science & technology studies (Actor–Network Theory, material political economy), and economic sociology (market-making, embeddedness).
Evidence used: Historical and contemporary examples and plausibility arguments (e.g., post-2022 funding shifts following ChatGPT, Microsoft–OpenAI partnership, role of Nvidia and cloud providers, data centers and energy infrastructures) to illustrate how decoys operate in practice.
Methodological orientation: Critical political-economy analysis that foregrounds network assemblages, infrastructural materialities (chips, data centers, energy), and financial flows rather than isolated technical artifacts.

Implications for AI Economics

Concentration and rent extraction: The Project of AI encourages vertical integration and control across the AI stack (chips, compute, models, platforms, distribution), increasing entry barriers and enabling new forms of economic rent.
Asset-heavy dynamics: Large fixed-cost investments (data centers, specialized chips, energy) amplify returns to scale and favor incumbents; economists should model these capital-intensity effects explicitly.
Finance and market-making matter: Financial actors, venture capital, strategic corporate partnerships, and lobbying shape incentives and technology trajectories; mapping capital flows is essential to understand realized outcomes.
Misplaced policy attention: Technical or limited regulatory fixes that do not address market structure, ownership, or infrastructure will at best mitigate symptoms and at worst legitimize incumbency.
Measurement priorities for researchers and policymakers:
- Track concentration of compute, ownership of models, and vertical integration across the stack.
- Trace financial relationships (investment, revenue-sharing, equity stakes) among cloud providers, model creators, and downstream firms.
- Quantify externalities (energy use, labor conditions in data labeling/annotation, geopolitics of supply chains).
Policy levers and interventions:
- Antitrust and competition policy targeted at vertical integration and platform gatekeeping.
- Public or cooperative provision of compute and data infrastructure to lower entry costs and democratize access.
- Conditional public funding (e.g., compute grants tied to access, governance, labor standards).
- Labor protections and standards for workers in AI supply chains (annotation, data center labor).
- Transparency requirements that go beyond model performance (financial ties, compute footprints, contract terms with cloud providers).
Research and advocacy strategy:
- Avoid being co-opted by decoys: orient research questions to networks, infrastructure, and economics rather than only to narrow technical diagnostics.
- Use interdisciplinary methods (political economy, network analysis, supply-chain tracing) to expose who benefits and how.
- Reframe public debate away from inevitability/ontology toward contestable choices about distribution, ownership, and governance.

Short recommendation distilled from the paper: hold actors to account by (a) detecting and resisting decoys that localize critique on technical particulars, and (b) interrogating and intervening on the material political economy—financing, infrastructure, ownership, and the networks that render AI power durable.

Assessment

Paper Typecommentary Evidence Strengthn/a — This is a conceptual and critical essay synthesizing literatures (communication, STS, economic sociology) rather than presenting original empirical analyses or causal tests, so it does not provide empirical evidence for causal claims. Methods Rigorn/a — No empirical methods are used; the paper relies on literature synthesis, conceptual argumentation, and normative analysis rather than systematic data collection or statistical identification strategies. SampleNo empirical sample; the paper is a world-building, conceptual analysis drawing on scholarly literatures in communication studies, science and technology studies, and economic sociology, and illustrative examples rather than original quantitative or qualitative data. Themesgovernance inequality GeneralizabilityConceptual arguments are not empirically validated and may not map directly onto specific industries, countries, or firm types, Potential Western/Global North bias in cited literature and examples, Does not provide testable hypotheses or precise mechanisms applicable to measurement in different contexts, Normative framing may limit applicability to positivist economic analyses or policymaking that requires quantifiable impacts

Claims (6)

Claim	Direction	Confidence	Outcome	Details
Those who fund and develop AI systems operate through and seek to sustain networks of power and wealth. Governance And Regulation	negative	high	operation and maintenance of networks of power and wealth by AI funders/developers	0.03
As AI funders and developers expand their access to resources and configure sociotechnical conditions, they benefit from decoys that animate scholars, critics, policymakers, journalists, and the public into co-constructing industry-empowering AI futures. Governance And Regulation	negative	high	co-construction of industry-empowering AI futures by multiple societal actors	0.03
Decoys often create the illusion of accountability while masking the emerging political economies that the Project of AI has set into motion. Governance And Regulation	negative	high	perceived accountability versus actual visibility of political economy	0.03
Decoys contribute to the network-making power that is at the heart of the Project's extraction and exploitation. Governance And Regulation	negative	high	network-making power and related extraction/exploitation	0.03
The authors identify five 'decoys' that seemingly critique—but in actuality co-constitute—AI's emergent power relations and material political economy. Governance And Regulation	negative	high	presence and role of five specific decoys in shaping AI power relations	0.01
Advancing meaningful fairness or accountability in AI requires: (1) recognizing when and how decoys serve as a distraction, and (2) grappling directly with the material political economy of the Project of AI. Governance And Regulation	positive	high	pathway/requirements to achieve meaningful fairness and accountability in AI	0.01