{"id":2175,"date":"2026-06-27T11:56:13","date_gmt":"2026-06-27T02:56:13","guid":{"rendered":"https:\/\/www.aicritique.org\/us\/?p=2175"},"modified":"2026-06-27T11:56:16","modified_gmt":"2026-06-27T02:56:16","slug":"could-ai-produce-a-corporate-feudal-order","status":"publish","type":"post","link":"https:\/\/www.aicritique.org\/us\/2026\/06\/27\/could-ai-produce-a-corporate-feudal-order\/","title":{"rendered":"Could AI Produce a Corporate Feudal Order"},"content":{"rendered":"\n

How platform power, AI infrastructure, private payment ecosystems, and weak fiscal capacity could push advanced capitalism toward something that resembles a new medievalism without ever fully becoming one<\/em><\/p>\n\n\n\n

Executive Summary<\/h2>\n\n\n\n

The strongest version of the \u201ccorporate feudalism\u201d thesis is not that technology companies will literally replace states, mint sovereign currencies, or own people in the medieval sense. It is that explosive technological innovation, especially in AI, may concentrate control over the decisive infrastructure of economic life, compute, cloud, chips, data, app distribution, payments, logistics, and algorithmic visibility, inside a relatively small number of firms. In that world, many businesses would not disappear, but would survive as dependent participants inside privately governed ecosystems whose owners set the tolls, the technical interfaces, the ranking rules, and the terms of continued access. The analogy to feudalism is imperfect, but the structural worry is real. Regulators in the United Kingdom, European Union, and United States have already described cloud, mobile, search, and AI markets in language of gatekeeping, lock-in, high switching costs, and entrenched concentration. <\/p>\n\n\n\n

The case for concern is strengthened by the sheer scale of the largest firms. In the current market snapshot, Nvidia was worth about $4.66 trillion, Microsoft about $3.70 trillion, Apple about $2.99 trillion, Amazon about $2.32 trillion, Alphabet about $2.10 trillion, and Meta about $1.83 trillion. That means individual technology firms now rival or exceed the annual output of major national economies such as Canada, France, and, in Microsoft\u2019s case, even the United Kingdom, though market capitalization is not the same thing as sovereign fiscal capacity. Their balance sheets are also enormous: Apple reported about $146.6 billion in cash, cash equivalents, and marketable securities in March 2026; Amazon about $143.1 billion; Alphabet about $126.8 billion; and Microsoft about $78.3 billion. <\/p>\n\n\n\n

At the same time, the thesis should not be overstated. History suggests that quasi-sovereign corporations often expand most when states are weak, fragmented, or willing to delegate power, and they often retreat when states regain administrative, military, fiscal, or regulatory capacity. Early modern company-states such as the East India Company and Dutch VOC were granted powers to collect taxes, administer justice, mint currency, and wage war; they were eventually constrained or absorbed as sovereign states became stronger. Today\u2019s big tech firms enjoy formidable power, but they face antitrust suits, gatekeeper regulation, AI law, export controls, data sovereignty rules, and rising geopolitical competition. In that sense, the most plausible future is not pure corporate sovereignty, but a contested terrain in which states and corporations co-govern key infrastructures, sometimes in rivalry and sometimes in partnership. <\/p>\n\n\n\n

The Economic Logic of Extreme Technological Concentration<\/h2>\n\n\n\n

AI is unusually hospitable to concentration because it sits atop multiple layers of infrastructure that are expensive, capital intensive, and hard to replicate. The OECD\u2019s 2025 competition roundtable on AI infrastructure describes the supply chain as running from chip inputs and energy networks through chip production, data centers, cloud platforms, and AI models to end users. It emphasizes high fixed costs, economies of scale, long lead times, technical complexity, and high barriers to entry across many levels of the chain. The same OECD paper also warns that several layers are already highly concentrated and that vertical and conglomerate integration is increasing as firms try to control multiple stages of the value chain. <\/p>\n\n\n\n

Cloud computing is a particularly important choke point because AI capability is increasingly inseparable from cloud access. Ofcom\u2019s cloud market study found that AWS and Microsoft together held an estimated 70% to 80% share of UK cloud infrastructure services in 2022, with Google the nearest challenger at 5% to 10%. Ofcom also found that egress fees, technical barriers, and interoperability problems can make switching harder and reinforce single-provider dependence. The CMA\u2019s 2025 cloud decision added that Microsoft and AWS had been earning sustained returns above their cost of capital and that their leading positions were likely to endure. Separately, the FTC\u2019s report on cloud-provider partnerships with leading AI developers warned that these deals can affect access to important inputs such as computing resources and engineering talent, can create lock-in effects, and can raise switching costs for AI developers themselves. <\/p>\n\n\n\n

Mobile ecosystems show the same logic in a more consumer-facing form. The UK CMA concluded that Apple and Google\u2019s duopoly gives them a \u201cstranglehold\u201d over crucial digital gateways, that they \u201cunilaterally determine the rules of the game,\u201d and that their decisions can be \u201cmake or break\u201d for thousands of businesses. It also found that Apple\u2019s App Store is the only permitted app store on iOS devices, that more than 90% of Android downloads occur through Google Play, and that both firms can use gatekeeper control to leverage power into adjacent markets. This is not territorial sovereignty, but it is a form of practical jurisdiction over access, visibility, pricing, and permissible conduct inside private domains that millions of firms and consumers cannot avoid. <\/p>\n\n\n\n

The practical reach of these firms into everyday life is staggering. Apple said in January 2026 that its installed base exceeded 2.5 billion active devices; by April it said the installed base had reached a new all-time high across major product categories and geographies. Meta reported 3.56 billion daily active people across its family of apps in March 2026. Alphabet reported that 2025 annual revenues exceeded $400 billion for the first time, with Google Cloud on an annual run rate above $70 billion and YouTube above $60 billion across ads and subscriptions. And in August 2024 a U.S. federal judge found that Google illegally maintained a search monopoly, with roughly 90% share in search and 95% on smartphones. This is influence over communication, discovery, authentication, work, consumption, advertising, and software distribution at a scale few states can match in ordinary daily experience. <\/p>\n\n\n\n

The historical parallels are therefore not frivolous. Early modern company-states complicated the line between state and corporation because they held monopolies, collected taxes, administered justice, minted currency, and used armed force while still being owned by investors and pursuing profit. Phillips and Sharman argue that these entities flourished when states were too weak or too administratively limited to govern distant territories directly and later declined as states strengthened. That history does not prove a return to company rule. But it does remind us that the modern separation between public sovereignty and private enterprise is historically contingent rather than natural or permanent. <\/p>\n\n\n\n

The Rise of Digital Vassalage<\/h2>\n\n\n\n

The most concrete evidence for a \u201ccorporate feudal\u201d dynamic lies not at the top of the system but in the middle. The relevant question is not whether small firms vanish, but whether they increasingly survive only by accepting dependence on platform owners who control market access. Recent academic work calls these actors \u201cplatform-dependent entrepreneurs\u201d: merchants, app developers, creators, hosts, drivers, and service providers who rely on platforms to reach customers, generate revenue, and obtain key resources. A 2024 systematic review found that platforms lower barriers to entry but also generate dependence and precarity, because governance shapes market access, visibility, and opportunity. A related Academy of Management Perspectives paper argues that power asymmetries are intrinsic to platform architecture and design, making entrepreneurial activity more precarious precisely because it is platform-dependent. <\/p>\n\n\n\n

For app developers, the \u201crules of the game\u201d are unusually literal. The CMA found that Apple and Google require certain developers to use in-app payment systems and can charge commissions of up to 30% on subscriptions and in-app digital purchases, while their control over app distribution lets them shape terms that businesses have \u201clittle choice but to accept.\u201d Apple\u2019s own developer materials still state that the standard commission on digital goods and services through the App Store is 30%, with a reduced 15% rate for firms in the Small Business Program. Google likewise applies a reduced 15% service fee on the first $1 million of annual earnings for enrolled developers, a concession that only underscores the baseline power of the store to define who pays what. <\/p>\n\n\n\n

For merchants and software businesses, the dependencies often stack. Amazon\u2019s fee schedules impose referral fees that vary by category, while sellers often face additional fulfillment charges if they rely on Fulfillment by Amazon. The FTC\u2019s monopolization suit against Amazon alleges that the company\u2019s conduct can overcharge sellers, degrade quality for shoppers, stifle innovation, and prevent rivals from competing fairly. And once firms rely on a hyperscale cloud provider, Ofcom warns that egress fees and technical barriers can make it costly to leave. The result is a multilayered dependency structure in which the same firm or set of firms can mediate storefront access, payments, fulfillment, advertising, and back-end infrastructure. <\/p>\n\n\n\n

For creators and gig workers, dependency often arrives through visibility and account governance rather than formal fees alone. Duffy and Meisner\u2019s research on creators describes workers as \u201cbeholden\u201d to \u201cinscrutable\u201d socio-technical systems in which algorithmic boosts, shadowbans, and moderation practices affect income and exposure. Uber\u2019s own materials define the service fee as the part of rider payments the firm retains after various deductions, and its Japanese rider help page explains that the platform deducts the service fee automatically from trip earnings. Airbnb\u2019s help center states that service and experience hosts typically pay fees of 15% and 20%, respectively, while Shopify\u2019s rules differentiate sharply between merchants who stay inside Shopify Payments and those who use third-party processors. In all of these cases, the platform is not merely matching supply and demand. It is governing the conditions of earning, ranking, and monetization. <\/p>\n\n\n\n

That is why the language of \u201cdigital tenancy\u201d is analytically useful even if it is rhetorically loaded. Medieval vassalage was sustained by asymmetries in access to the essential asset of the age, land. Today\u2019s equivalent assets are not plots of earth but user demand, default placement, data, payment rails, cloud services, and algorithmic attention. Platform owners rarely command by direct coercion. They rule more subtly, by contract, API, policy update, ranking signal, visibility threshold, moderation standard, and deplatforming risk. Yet for many dependent firms, the lived experience is similar to inhabiting another authority\u2019s jurisdiction without meaningful voice in the lawmaking process. <\/p>\n\n\n\n

AI, Labor, and the Fiscal Problem of Redistribution<\/h2>\n\n\n\n

If the corporate-feudal scenario has a political economy core, it is this: what happens when productivity and profit rise faster than labor income, but the state cannot effectively redistribute the gains? The current evidence on AI and employment is more mixed than some popular narratives suggest. A 2025 NBER paper found that tasks with higher AI exposure subsequently face reduced labor demand, but also that productivity effects and task reallocation can offset some job losses at the firm level. In June 2026, an ECB study reported that the aggregate employment and wage impact in the United States had so far been muted, even though highly exposed occupations declined and low-substitution occupations expanded. Brookings has similarly argued that policymakers should plan for displacement without assuming a simple one-way march to mass unemployment. <\/p>\n\n\n\n

Even so, the distributional concern is serious. The IMF argues that the AI transition will require stronger social safety nets, investment in education, and tax systems that support human workers and mitigate inequality. It recommends more generous and portable unemployment insurance, possible wage insurance, sector-based training, and broader social assistance. The IMF also argues that AI may increase wealth concentration and that taxes on capital income, stronger corporate taxation, and better enforcement may be needed to preserve the tax base and finance broader sharing of the gains. Related IMF research has explicitly warned that AI may reduce wage inequality in some cases while substantially increasing wealth inequality because ownership of capital and intangible assets becomes more decisive. <\/p>\n\n\n\n

This is where debates over UBI, negative income tax, wage subsidies, and social insurance become relevant. The OECD\u2019s long-standing position is cautious: more universal income support may become necessary in stages, but financing is the hard question. Brookings has argued that a true UBI is often too expensive and poorly targeted relative to more specific benefits or a negative income tax design. These are not anti-redistribution arguments; they are institutional arguments. They acknowledge that if labor income weakens relative to AI-driven rents, governments will face pressure to build transfer systems, but also that universal cash promises are hard to sustain without an effective means of taxing the winners. <\/p>\n\n\n\n

And that is precisely the problem. The OECD\u2019s BEPS project exists because multinational firms can exploit gaps and mismatches in tax rules to shift profits and avoid taxation, especially where value is tied to intangibles rather than easily located physical assets. The 2021 two-pillar tax agreement was meant to modernize this system, yet Pillar One remains unresolved, digital services taxes remain politically contentious, and major powers still retaliate when market jurisdictions try to tax digital giants directly. Canada rescinded its digital services tax in 2025 to revive trade negotiations with the United States, and the U.S. administration in June 2026 threatened 100% tariffs on countries imposing such taxes. The result is not that taxation is impossible, but that it is geopolitically fraught, administratively difficult, and vulnerable to jurisdictional arbitrage. <\/p>\n\n\n\n

This is the hinge on which the darker scenario turns. If states cannot reliably tax concentrated AI rents, then the welfare state becomes harder to finance just as its need may be increasing. That does not automatically produce corporate feudalism. But it does raise the possibility that private actors, not states, will become the more effective distributors of marginal purchasing power, whether through loyalty credits, subsidized subscriptions, ecosystem wallets, or platform-specific benefits designed to preserve demand inside their own domains. <\/p>\n\n\n\n

Corporate Credits as a Private Form of Basic Income<\/h2>\n\n\n\n

The idea of corporations distributing a kind of private basic income sounds futuristic until one notices how much of it already exists in fragmented form. Amazon issues promotional and courtesy credits. Apple maintains account balances that can be funded through gift cards and then spent within Apple\u2019s commerce environment. Uber operates Uber Cash and Uber Credit, which can be applied to orders and rides within that ecosystem and are denominated in the same currency as the user\u2019s Uber activity. Roblox operates Robux as a virtual currency that users buy for real money or earn as creators, while the company explicitly states that Robux itself has no value in real currency except under company-defined exchange programs for eligible developers. These are not welfare programs. But they are company-issued units of purchasing power, often revocable, conditional, or ecosystem-bound. <\/p>\n\n\n\n

Academic work on points and platform money suggests that these systems are more economically significant than they appear. Yuri Okina\u2019s analysis of Japanese cashless payments argues that loyalty points used by platform-style companies increasingly function like \u201ccorporate currencies\u201d or a \u201cpseudo-currency,\u201d especially when they circulate inside point-based economic zones created through partnerships with member stores. The BIS paper \u201cCompeting Digital Monies\u201d goes further by formally modeling competition among bank deposits, digital platform tokens, and CBDCs, and notes that big-tech payment instruments already operate as closed-loop systems within specific ecosystems. This matters because money-like instruments need not be legal tender to shape behavior; they only need to be accepted within the places where people actually consume. <\/p>\n\n\n\n

The issuer\u2019s incentives are also revealing. A 2024 NBER working paper on loyalty tokens argues that such arrangements generate switching costs and examines their role as redeemable platform currencies. Its analysis finds that issuers optimally prefer non-tradable tokens and explains that non-tradability can serve as a financing tool by increasing front-loaded cash flow. Okina similarly emphasizes that point systems remain most effective for issuing firms when they operate in somewhat closed networks that exclude competitors. In other words, the economic logic of private purchasing power is not generosity but enclosure: maintain demand, increase retention, gather data, reduce churn, and prevent exit. <\/p>\n\n\n\n

That is why the phrase \u201ccorporate basic income\u201d should be treated as an analytic provocation, not a description of current policy. A real UBI is universal, portable, and state-backed. Corporate credits are partial, conditional, and proprietary. But under conditions of weak wage growth and weak state redistribution, firms may find it rational to subsidize consumption directly within their ecosystems rather than surrender more revenue through general taxation. The resulting system would support demand while deepening dependence. Purchasing power would become more plentiful in nominal terms but less free in social terms, because it could only be spent through particular rails, on particular terms, under rules set by private authorities. <\/p>\n\n\n\n

This is not hypothetical in a narrow sense. Airline loyalty programs have become large and credible enough to serve as financing collateral, illustrating that \u201cpoints\u201d and proprietary rewards can become major balance-sheet assets. During the past several years, airlines repeatedly borrowed against loyalty programs, and in 2024 JetBlue raised financing partly secured by TrueBlue. The broader lesson is not about aviation in particular. It is that company-issued claims on future consumption now have enough durability, predictability, and market value to function as serious economic instruments. In a more AI-concentrated economy, one can imagine the same logic spreading far beyond travel. <\/p>\n\n\n\n

What the Medieval Analogy Clarifies and What It Obscures<\/h2>\n\n\n\n

Used carefully, the medieval analogy helps reveal structural similarities that ordinary market language can hide.<\/p>\n\n\n\n

Land and fiefs become data, compute, cloud, distribution channels, and payment rails.<\/strong> In feudal orders, control over land determined who could produce and survive. In digital economies, the decisive productive assets are often proprietary infrastructures and data-rich gateways. Businesses do not need title deeds from Big Tech, but they increasingly need permissioned access to clouds, app stores, ranking systems, wallets, ad markets, and APIs. <\/p>\n\n\n\n

Lords become platform owners and AI infrastructure firms.<\/strong> Like lords, dominant platforms can levy recurring tolls, define conditions of access, and extract tribute from activity conducted on \u201ctheir\u201d terrain. The CMA\u2019s finding that Apple and Google determine the \u201crules of the game,\u201d Ofcom\u2019s concern over cloud switching barriers, and the FTC\u2019s concern over input foreclosure in AI partnerships all fit that pattern. <\/p>\n\n\n\n

Vassals become dependent businesses and workers.<\/strong> Merchants, drivers, hosts, sellers, developers, and creators can retain formal independence while depending on a superior power for market access and revenue. The platform-dependence literature explicitly frames this relationship in terms of asymmetry, precarity, and limited countervailing power. <\/p>\n\n\n\n

Serfdom becomes lock-in.<\/strong> Modern consumers are not legally unfree, but they can become practically captive through defaults, network effects, ecosystem bundles, data portability frictions, loyalty programs, and accumulated digital assets that are costly to abandon. This is especially visible in app ecosystems, cloud migrations, and point-based payment environments. <\/p>\n\n\n\n

Rents and tithes become commissions, subscriptions, ads, and data extraction.<\/strong> The defining income form in feudal analogies is not productive profit alone but recurring claims on others\u2019 activity. Up-to-30% in-app commissions, marketplace fees, service fees, switching costs, and monetized behavioral data all resemble privatized toll roads more than textbook competitive exchange. <\/p>\n\n\n\n

Jurisdictions become ecosystems.<\/strong> Medieval power was patchy and layered, not neatly territorial. Private digital ecosystems are similarly jurisdictional: they have rules, sanctions, rights of passage, and internal currencies, but their boundaries are functional rather than geographic. <\/p>\n\n\n\n

Still, the analogy can also mislead. Modern technology firms remain capitalist corporations with shareholders, financial reporting, labor costs, and competitive pressures; they do not enjoy a lawful monopoly on violence, nor do they possess the full tax-and-police sovereignty of states. Jeremy Gilbert argues that \u201ctechno-feudalism\u201d is better understood as a new regime of capitalist accumulation rather than a wholly new mode of production, and Nick Gane similarly argues that capitalism has not been superseded so much as intensified in digital form. The medieval frame is most useful when it identifies rent extraction, hierarchy, enclosure, and private rule-making; it becomes less useful when it implies that markets, wages, democratic politics, and state coercion have disappeared. <\/p>\n\n\n\n

History also offers a built-in warning against fatalism. Company-states flourished under conditions of weak public capacity and later declined as sovereign states professionalized taxation, administration, and power projection. Phillips and Sharman\u2019s account is therefore a direct challenge to deterministic stories about corporate takeover: private quasi-sovereignty expands when states choose to delegate, cannot coordinate, or lack capacity. It contracts when states decide that the private intermediary has become too strategically important to remain largely uncontrolled. <\/p>\n\n\n\n

State Power, Counterarguments, and Alternative Futures<\/h2>\n\n\n\n

There are strong reasons to resist the claim that a corporate-feudal order is inevitable. First, states are already pushing back. The European Commission designated Alphabet, Amazon, Apple, ByteDance, Meta, and Microsoft as DMA gatekeepers, and it has already fined Apple and Meta for non-compliance. The EU AI Act entered into force in 2024 and is becoming fully applicable in stages through 2026 and beyond. In the United States, Google has already lost a landmark search monopolization case. In the UK, cloud concentration has been formally investigated, and lawmakers have warned that overreliance on Palantir in the public sector is an \u201cunacceptable point of weakness.\u201d These are not signs of state surrender. They are signs of a slow, uneven reassertion of sovereignty. <\/p>\n\n\n\n

Second, states can also become more powerful by becoming more digital. Central banks continue to explore CBDCs, which are in part a response to the rise of private payment ecosystems. BIS survey results indicate that 91% of the 93 surveyed central banks were exploring retail or wholesale CBDCs in 2024, with wholesale work generally more advanced. Meanwhile, public-private partnerships in defense and research show not simply corporate dominance but state-corporate fusion: the Pentagon\u2019s JWCC cloud contract was awarded to Amazon, Microsoft, Google, and Oracle, while governments increasingly seek domestic or sovereign alternatives in AI and cloud to reduce strategic dependence. That is not medieval fragmentation. It is the emergence of hybrid governance. <\/p>\n\n\n\n

Third, the technology itself may not concentrate as much as today\u2019s headlines imply. Stanford\u2019s 2025 AI Index found that open-weight models sharply narrowed the performance gap with closed models and that the inference cost of GPT-3.5-level capability fell more than 280-fold between late 2022 and late 2024. JPMorganChase Institute data show that the monthly cost of AI tools for small businesses fell from around $50 in 2019 to $20 to $30 in 2025. OECD work on AI diffusion shows that adoption gaps are widening in some sectors and regions, but it also documents rapid diffusion across countries and firms. These trends support a real counter-thesis: AI may lower barriers to entrepreneurship for many firms even while concentrating rents at the infrastructure layer. <\/p>\n\n\n\n

Fourth, political economy outcomes depend on institutions, not technology alone. If public cloud procurement promotes interoperability, if competition policy restrains self-preferencing, if labor law expands to capture platform work, if open models remain viable, and if states build public digital infrastructure, then concentration can be contained. Brookings points to India\u2019s subsidized public compute pool as one possible way to democratize access to AI infrastructure. The OECD\u2019s AI openness work likewise suggests that openness can produce benefits alongside risks, and that policy should focus on calibrating governance rather than simply defending incumbent closed systems. <\/p>\n\n\n\n

The most plausible futures therefore look less like prophecy than branching scenarios.<\/p>\n\n\n\n

Regulated platform capitalism.<\/strong> Under this outcome, dominant firms remain powerful, but antitrust, interoperability mandates, cloud portability rules, AI regulation, data protection, and tax reform prevent them from becoming quasi-sovereign. Small firms still depend on platforms, but dependence is moderated by enforceable rights, contestability, and public oversight. This is the logic behind the DMA, AI Act, the Google antitrust cases, and current cloud investigations. <\/p>\n\n\n\n

Corporate feudalism.<\/strong> Under this darker outcome, AI-driven rents concentrate faster than states can respond; cloud, chips, payments, and attention become locked into a handful of vertically integrated ecosystems; labor bargaining power weakens; and redistribution shifts from public cash to private credits, subscriptions, and loyalty systems that preserve demand while deepening enclosure. The formal shell of citizenship remains, but practical dependence migrates toward corporate jurisdictions. This outcome is not yet here, but the underlying mechanisms are visible. <\/p>\n\n\n\n

State-corporate hybrid governance.<\/strong> This may be the likeliest near-term path. Governments increasingly regulate, subsidize, and depend on major technology firms at the same time. Defense clouds, public-sector analytics, AI procurement, and national-security controls create relationships that are neither simple domination nor simple regulation. Sovereignty is not lost so much as co-produced through contracts, technical standards, and strategic dependencies. <\/p>\n\n\n\n

Decentralized and open technological pluralism.<\/strong> In this more optimistic future, open-weight AI, cheaper inference, public compute pools, cooperative platforms, and competition among jurisdictions prevent any stable digital aristocracy from emerging. Individuals and SMEs gain real leverage because the cost of capability falls faster than the cost of switching. The current evidence does not guarantee this outcome, but it keeps it firmly within the realm of possibility. <\/p>\n\n\n\n

The central question, then, is not whether AI will magically turn the twenty-first century into the twelfth. It is whether societies will allow ownership of the essential infrastructures of intelligence, commerce, communication, and payment to harden into a system of private rule so entrenched that market dependence starts to resemble political dependence. On current evidence, that possibility is real enough to take seriously, but far from inevitable. The future will hinge less on technology itself than on law, taxation, interoperability, labor power, public capacity, and the political willingness to treat digital infrastructure as a site of constitutional importance rather than just another profitable industry. <\/p>\n\n\n\n

Bibliography<\/h2>\n\n\n\n
    \n
  • OECD,\u00a0Competition in Artificial Intelligence Infrastructure<\/em>\u00a0(2025).\u00a0<\/li>\n\n\n\n
  • Ofcom,\u00a0Cloud Services Market Study: Final Report<\/em>\u00a0(2023).\u00a0<\/li>\n\n\n\n
  • UK Competition and Markets Authority,\u00a0Mobile Ecosystems Final Report Summary<\/em>\u00a0(2022).\u00a0<\/li>\n\n\n\n
  • U.S. Federal Trade Commission,\u00a0Partnerships Between Cloud Service Providers and AI Developers<\/em>\u00a0(2025).\u00a0<\/li>\n\n\n\n
  • U.S. Federal Trade Commission,\u00a0FTC Sues Amazon for Illegally Maintaining Monopoly Power<\/em>\u00a0(2023).\u00a0<\/li>\n\n\n\n
  • IMF Blog,\u00a0Fiscal Policy Can Help Broaden the Gains of AI to Humanity<\/em>\u00a0(2024).\u00a0<\/li>\n\n\n\n
  • OECD,\u00a0Basic Income as a Policy Option<\/em>\u00a0(2017).\u00a0<\/li>\n\n\n\n
  • Brookings Institution,\u00a0Universal Basic Income as a Policy Response to Current Challenges<\/em>\u00a0(2019).\u00a0<\/li>\n\n\n\n
  • OECD,\u00a0Base Erosion and Profit Shifting<\/em>\u00a0overview and\u00a0Tax Challenges Arising from Digitalisation of the Economy<\/em>\u00a0materials.\u00a0<\/li>\n\n\n\n
  • Bank for International Settlements,\u00a0Competing Digital Monies<\/em>\u00a0(2025).\u00a0<\/li>\n\n\n\n
  • Yuri Okina,\u00a0Digitalization of Payment Instruments: Cashless Payments and Loyalty Points Systems<\/em>\u00a0(2022).\u00a0<\/li>\n\n\n\n
  • NBER,\u00a0Loyalty Tokens<\/em>\u00a0working paper (2024).\u00a0<\/li>\n\n\n\n
  • Songping Yu and Tomoki Sekiguchi,\u00a0Platform-Dependent Entrepreneurship: A Systematic Review<\/em>\u00a0(2024).\u00a0<\/li>\n\n\n\n
  • Duffy and Meisner,\u00a0Platform Governance at the Margins: Social Media Creators\u2019 Experiences with Algorithmic (In)Visibility<\/em>\u00a0(2022).\u00a0<\/li>\n\n\n\n
  • Cutolo and Kenney,\u00a0Platform-Dependent Entrepreneurs: Power Asymmetries, Risk, and Strategy in the Platform Economy<\/em>\u00a0(2019).\u00a0<\/li>\n\n\n\n
  • Menaka Hampole, Dimitris Papanikolaou, Lawrence D.W. Schmidt, and Bryan Seegmiller,\u00a0Artificial Intelligence and the Labor Market<\/em>\u00a0(NBER, 2025).\u00a0<\/li>\n\n\n\n
  • Stanford HAI,\u00a0AI Index Report 2025<\/em>.\u00a0<\/li>\n\n\n\n
  • JPMorganChase Institute,\u00a0Understanding the Use of AI Among Small Businesses<\/em>\u00a0(2026).\u00a0<\/li>\n\n\n\n
  • OECD,\u00a0Emerging Divides in the Transition to Artificial Intelligence<\/em>\u00a0(2025).\u00a0<\/li>\n\n\n\n
  • OECD,\u00a0AI Openness: A Primer for Policymakers<\/em>\u00a0(2025).\u00a0<\/li>\n\n\n\n
  • Andrew Phillips and J.C. Sharman,\u00a0Company-States and the Creation of the Global International System<\/em>\u00a0(2020).\u00a0<\/li>\n\n\n\n
  • Philip J. Stern,\u00a0The Company-State: Corporate Sovereignty and the Early Modern Foundations of the British Empire in India<\/em>\u00a0(2011).\u00a0<\/li>\n\n\n\n
  • Jeremy Gilbert,\u00a0Techno-Feudalism or Platform Capitalism? Conceptualising the Digital Society<\/em>\u00a0(2024).\u00a0<\/li>\n\n\n\n
  • Nick Gane,\u00a0Capitalism Is Capitalism, Not Technofeudalism<\/em>\u00a0(2025).\u00a0<\/li>\n\n\n\n
  • European Commission, Digital Markets Act gatekeeper materials and non-compliance enforcement.\u00a0<\/li>\n\n\n\n
  • European Commission, AI Act implementation materials.\u00a0<\/li>\n\n\n\n
  • Reuters reporting on Google\u2019s search monopoly ruling, digital-services-tax conflicts, AWS\/Azure gatekeeper scrutiny, public-sector dependence, and defense\/public-cloud contracting.\u00a0<\/li>\n\n\n\n
  • Corporate filings and investor materials from Apple, Amazon, Alphabet, Microsoft, and Meta for user metrics, cash reserves, and operating scale.\u00a0<\/li>\n<\/ul>\n","protected":false},"excerpt":{"rendered":"

    How platform power, AI infrastructure, private payment ecosystems, and weak fiscal capacity could push advanced capitalism toward something that resembles a new medievalism without ever fully becoming one Executive Summary The strongest version of the \u201ccorporate feudalism\u201d thesis is not…<\/p>\n","protected":false},"author":4,"featured_media":2176,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[69,10,21,34,71,19,11],"tags":[],"class_list":["post-2175","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-ai-governance","category-employment","category-main","category-nation","category-philosophy-of-ai","category-regulation","category-society"],"_links":{"self":[{"href":"https:\/\/www.aicritique.org\/us\/wp-json\/wp\/v2\/posts\/2175","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.aicritique.org\/us\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.aicritique.org\/us\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.aicritique.org\/us\/wp-json\/wp\/v2\/users\/4"}],"replies":[{"embeddable":true,"href":"https:\/\/www.aicritique.org\/us\/wp-json\/wp\/v2\/comments?post=2175"}],"version-history":[{"count":1,"href":"https:\/\/www.aicritique.org\/us\/wp-json\/wp\/v2\/posts\/2175\/revisions"}],"predecessor-version":[{"id":2177,"href":"https:\/\/www.aicritique.org\/us\/wp-json\/wp\/v2\/posts\/2175\/revisions\/2177"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.aicritique.org\/us\/wp-json\/wp\/v2\/media\/2176"}],"wp:attachment":[{"href":"https:\/\/www.aicritique.org\/us\/wp-json\/wp\/v2\/media?parent=2175"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.aicritique.org\/us\/wp-json\/wp\/v2\/categories?post=2175"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.aicritique.org\/us\/wp-json\/wp\/v2\/tags?post=2175"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}