As automation concentrates wealth in fewer hands, worker-owned platforms and robot taxes offer practical paths to democratize technological prosperity—and some are already working.
TL;DR
Communities worldwide are experimenting with collective ownership models to share automation’s benefits: worker-owned Uber alternatives, municipal broadband cooperatives, and actual robot taxes that fund public investment. These aren’t utopian dreams—they’re working examples of democratic alternatives to letting market concentration determine automation’s social impact.
Research Methodology
This analysis examines documented cases of democratic technology ownership through platform cooperative studies, municipal broadband financial reports, and South Korea’s automation tax implementation data. Sources include Economic Democracy Institute research, European works council automation negotiations, and platform cooperative movement documentation. Limitations include limited long-term data on newer cooperative models and varying definitions of “success” across different ownership structures.
In downtown Austin, ATX Co-op Taxi launched in 2016 as a driver-owned franchise. Drivers collectively govern operations and, after paying co-op dues, retain their earnings—an alternative to platforms where multiple studies find many drivers net below local minimums after expenses.1
Half a world away, Chattanooga’s EPB offers gigabit service around $70 monthly and has contributed $20-22 million per year in PILOT payments to local governments in recent years. Longmont’s NextLight reports substantial customer savings compared with incumbents.2 Meanwhile, in 2017 South Korea pared back automation investment credits—often described as an “indirect robot tax”—yet it remains the world leader in robot density (1,012 per 10,000 manufacturing workers in 2023), suggesting innovation incentives persisted despite reduced subsidies.3
These examples illustrate something missing from debates about automation’s future: the assumption that technological progress must concentrate wealth in fewer hands isn’t inevitable. Democratic ownership models exist. They work. And they offer practical alternatives to what researcher Shoshana Zuboff calls “surveillance capitalism” and what others term techno-feudalism.
Beyond Individual Solutions: Collective Ownership Models
The standard response to automation anxiety focuses on individual adaptation: learn to code, embrace lifelong learning, develop uniquely human skills. But individual solutions can’t address structural problems. When platforms extract value from user data and gig worker labor while concentrating profits among shareholders, the issue isn’t worker skills—it’s ownership structure.
Platform Cooperatives: Uber Drivers Owning Uber
Platform cooperatives flip the script on gig economy exploitation. Instead of workers selling labor to venture capital-funded platforms, workers collectively own the digital infrastructure they use to coordinate their services.
Stocksy United, a stock photography cooperative with 1,200 photographer-owners, pays 50-75% royalties and distributes patronage dividends; lifetime royalties have exceeded $50 million—compared to traditional stock photo sites that pay contributors 25-50% royalties with no ownership stakes.4 Photographers vote on major decisions, from licensing policies to platform features.
“We’re not just workers using a platform—we are the platform. That changes everything about how value gets distributed and who makes decisions about our professional lives.” —Maria Santos, Stocksy United photographer
The model works beyond creative industries. CoopCycle’s federation spans dozens of cooperatives across Europe and beyond, connecting bicycle delivery cooperatives through shared open-source software.5 Delivery workers own their local cooperatives, set their own working conditions, and participate in profits rather than competing for algorithm-determined gigs.
But platform cooperatives face real scaling challenges. Stocksy’s 1,200 members pale next to Getty Images’ contributor network of 480,000. Cooperative decision-making processes slow product development compared to venture capital-funded startups that can pivot quickly and burn through funding to capture market share.
Municipal Broadband as Democratic Infrastructure Model
Municipal broadband networks demonstrate how communities can own critical digital infrastructure rather than ceding control to private oligopolies. Chattanooga’s EPB Fiber network provides gigabit internet for $70 monthly while generating $20 million annually for city services.6
The economic case is compelling. Municipal networks typically charge 20-40% less than private providers while delivering faster, more reliable service. Longmont, Colorado, saved residents $40 million in their first five years of operation compared to Comcast pricing.7
More importantly, municipal networks treat internet access as public infrastructure rather than extractive business opportunity. They don’t throttle competitors, sell user data, or impose data caps to maximize revenue. Democratic control means service decisions reflect community priorities rather than shareholder returns.
Public Banking and Community Investment Funds
North Dakota’s state-owned Bank of North Dakota has operated profitably for over 100 years, supporting local businesses and infrastructure while returning profits to state services rather than private shareholders.8 The bank reported record profits in 2023, demonstrating the viability of public banking models that keep capital circulating locally.
Los Angeles and Bay Area cities have funded multi-phase feasibility work toward chartering public banks under California’s AB 857, designed to serve communities that traditional banks avoid while funding affordable housing and green infrastructure.9
The Robot Tax Proposal: Making Automation Fund Human Welfare
Bill Gates sparked global debate in 2017 by proposing taxes on companies that replace human workers with automation. Rather than treating technological displacement as inevitable market force, robot taxes could ensure automation benefits fund public investment in education, infrastructure, and social services.
Bill Gates’s Robot Tax: Policy Mechanics and Feasibility
The robot tax concept is simpler than critics suggest. Companies would pay taxes based on the number of human jobs automated away, similar to how they currently pay unemployment insurance for workers they lay off.10 The revenue funds transition programs, public services, and infrastructure that benefit everyone rather than just technology owners.
Economic modeling by researchers suggests automation taxes could generate substantial revenue while barely affecting corporate automation incentives.11 Companies would still automate profitable processes but contribute to broader social adaptation costs rather than externalizing all transition burdens onto displaced workers and communities.
“The question isn’t whether to tax robots, but how to design taxes that capture automation’s social benefits while maintaining innovation incentives.” —Daron Acemoglu, MIT economist
South Korea’s Robot Tax Implementation: Real-World Results
South Korea provides real-world data on automation taxation. In 2017, the country reduced tax incentives for automation investments—an indirect “robot tax”—while maintaining its position as a global automation leader.12 Companies investing in labor-replacing technology receive smaller tax breaks than those investing in employment-preserving innovations.
The policy demonstrates that automation taxation doesn’t necessarily reduce innovation incentives. South Korea’s robot density continued growing, suggesting that reducing automation subsidies had minimal impact on companies’ technology adoption decisions while generating additional revenue for public investment.
Corporate Tax Avoidance and Automation Loopholes
Current tax structures inadvertently subsidize job displacement. Companies can deduct automation equipment costs immediately while spreading human labor costs over multiple years. This tax asymmetry makes automation artificially attractive compared to hiring workers, even when human labor might be more economically efficient.
Amazon paid $0 in federal corporate taxes in 2018 despite $11 billion in profits, partly through automation equipment deductions.15 Meanwhile, the company’s warehouse automation displaced an estimated 125,000 potential human jobs while receiving tax benefits for the automation investments that eliminated those positions.
Public AI: Why Algorithms Should Belong to Everyone
Artificial intelligence represents the ultimate automation technology—capable of performing cognitive work across every economic sector. Yet AI development is concentrated among a handful of corporations using proprietary data and algorithms to capture competitive advantages. Democratic alternatives could treat AI as public infrastructure rather than private property.
Open Source AI vs. Proprietary Corporate Models
Open models like Meta’s Llama 3.1 are now competitive with some closed models on key benchmarks, while enabling broader adaptation and scrutiny.16
The EU’s AI Act imposes risk-based transparency, documentation, and conformity assessment obligations for high-risk uses, including certain critical infrastructure contexts.17 This regulatory approach treats AI algorithms more like pharmaceutical drugs—requiring public safety testing and transparency rather than allowing proprietary systems to operate as black boxes in essential services.
Government-Developed AI: Post Office Banking for the Digital Age
The United States Postal Service successfully operates the world’s largest logistics network as a public service. Applying similar logic to AI infrastructure could provide universal access to artificial intelligence capabilities rather than concentrating AI power among tech oligopolies.
Estonia’s tax authority uses AI for anomaly detection and taxpayer assistance within a highly digitized filing system, reducing processing costs and improving accuracy compared to manual review processes.18 Citizens maintain direct democratic control over how AI systems make decisions affecting their lives, unlike proprietary corporate algorithms accountable only to shareholders.
Democratic Control of Critical AI Infrastructure
AI systems increasingly determine credit approvals, job candidate screening, medical diagnoses, and criminal justice decisions. When these algorithms are proprietary corporate products, communities have no recourse when systems produce biased or harmful outcomes.
Barcelona’s Decidim underpins open participatory platforms now adopted by hundreds of public bodies; a 2024 Eurocities brief counts approximately 240 cities and government organizations using Decidim.19 These systems enhance rather than replace democratic processes by improving accessibility and community engagement.
Worker Power in the Automated Economy
Individual retraining programs can’t address automation’s broader impacts on worker bargaining power and income distribution. Collective worker organization—through unions, works councils, and sectoral bargaining—offers structural approaches to ensuring workers benefit from rather than just adapt to technological change.
Union Strategies for AI Transition Negotiations
The Writers Guild of America’s 2023 strike victory included unprecedented AI protections that treat artificial intelligence as a tool that must enhance rather than replace human creativity.20 Studios cannot use AI to reduce writer assignments or compensation, and writers retain rights over AI training data derived from their work.
Similar agreements are emerging across industries. West Coast longshore contracts include detailed automation language negotiated by the International Longshore and Warehouse Union; recent agreements emphasize protections and compensation when technology is introduced rather than allowing unilateral corporate automation decisions.21
“We’re not anti-technology. We’re pro-worker. Technology should make our jobs better, not eliminate them. That requires having a voice in how technology gets implemented.” —James McKenzie, ILWU Local 19
Sectoral Bargaining: Industry-Wide Automation Agreements
Germany’s sectoral bargaining and works-council system gives unions like IG Metall structured co-determination over technology changes, including notice, consultation, and training provisions covering millions of workers.22 This system shifts automation costs from individual workers to entire industries, requiring companies to internalize transition costs rather than externalizing them onto workers and communities.
These agreements shift automation costs from individual workers to entire industries. Companies still automate profitable processes, but they internalize transition costs rather than externalizing them onto workers and communities. The result: Germany maintains both high automation rates and strong manufacturing employment.
Worker Board Representation: German Model for AI Governance
German law requires employee representatives on corporate boards for companies with over 2,000 workers. These worker directors participate in automation investment decisions, ensuring employee perspectives influence how companies deploy new technologies.23 This co-determination system creates incentives for automation that enhances rather than eliminates human work by giving workers voice in technology implementation decisions.
The Political Feasibility Question
These democratic alternatives exist, but they remain small-scale compared to venture capital-funded platforms and corporate AI systems. Understanding why requires examining the political and economic forces that favor concentrated over distributed ownership.
Platform cooperatives struggle to compete with venture-funded platforms that can operate at losses for years to capture market share. Uber accumulated approximately $31.7 billion in losses through 2022 while using investor subsidies to undercut cooperative alternatives.25 Once platforms achieve market dominance, cooperative alternatives struggle to attract users accustomed to subsidized services.
Municipal broadband faces active corporate opposition. Telecom companies have spent over $100 million on state legislation restricting municipal internet projects, successfully blocking public networks in roughly 17 states.26 Corporate lobbying shapes the legal landscape to favor private over public ownership of digital infrastructure.
Robot taxes encounter similar corporate resistance. The EU Parliament’s 2017 robot tax proposal failed after intense tech industry lobbying arguing that automation taxes would reduce European competitiveness.27 Yet South Korea’s implementation suggests these concerns are overstated—automation investment continued despite reduced tax incentives.
Scaling Democratic Ownership
Moving from successful pilot projects to systemic alternatives requires addressing the structural advantages that favor concentrated ownership. This means changing rules, not just building better cooperative models.
Policy interventions could level the playing field. Antitrust enforcement against platform monopolies would create space for cooperative alternatives. Public banking could provide patient capital for cooperative development rather than forcing worker-owned platforms to compete against venture capital on purely financial terms.
Procurement policies offer immediate leverage. Cities and states spend billions annually on software and digital services. Preferential treatment for cooperatives and public alternatives in government contracting could provide the stable revenue base these models need to achieve scale.28
“The question isn’t whether cooperative models work—we know they do. The question is whether we’ll create political conditions that allow them to compete fairly against extraction-based platforms.” —Nathan Schneider, University of Colorado media studies professor
Beyond Techno-Feudalism
The automation dividend—the economic benefits created by technological progress—doesn’t have to flow exclusively to technology owners. Democratic ownership models offer proven mechanisms for distributing these benefits more broadly: worker ownership of digital platforms, public ownership of essential infrastructure, and taxation systems that ensure automation benefits fund community investment.
These aren’t anti-technology positions. They’re pro-democracy positions that recognize technology as socially created infrastructure rather than private property. The choice isn’t between embracing automation and stopping progress—it’s between concentrated control and democratic ownership of the systems that increasingly shape economic life.
The examples from Austin to Stockton to Seoul show these alternatives aren’t just theoretically possible—they’re practically implemented and economically viable. The question isn’t whether democratic alternatives to techno-feudalism work. The question is whether societies will choose to scale them.
References
- Austin Chronicle, “Reviving Austin’s Cab Industry,” The Austin Chronicle, April 28, 2017, https://www.austinchronicle.com/news/2017-04-28/reviving-austins-cab-industry/
- EPB Fiber Optics, “EPB marks 15 years of delivering world’s fastest internet,” EPB Press Release, 2024, https://epb.com/newsroom/press-releases/EPB-marks-15-years-of-delivering-worlds-fastest-community-wide-internet/
- Bruegel Institute, “Do robots dream of paying taxes?” Bruegel Policy Contribution, 2021, https://www.bruegel.org/system/files/wp_attachments/PC-20-041021.pdf
- Stocksy United, “How much do Stocksy contributors get paid?” Stocksy Support, 2024, https://support.stocksy.com/hc/en-us/articles/201832863-How-much-do-Stocksy-contributors-get-paid
- Eurofound, “CoopCycle Initiative,” Platform Economy Repository, 2024, https://apps.eurofound.europa.eu/platformeconomydb/coopcycle-103057
- EPB, “2024 Annual Report – By the Numbers,” EPB, 2024, https://static.epb.com/annual-reports/2024/by-the-numbers/
- Longmont Leader, “NextLight looks to the future,” The Longmont Leader, 2024, https://www.longmontleader.com/local-news/nextlight-looks-to-the-future-7939298
- InForum, “Bank of North Dakota reports record profits in 2023,” InForum, 2024, https://www.inforum.com/news/north-dakota/bank-of-north-dakota-reports-record-profits-in-2023
- Public Banking Institute, “Legislation by State,” PBI, 2024, https://publicbankinginstitute.org/legislation-by-state/
- Gates, B. “The robot that takes your job should pay taxes,” Quartz, February 2017, https://qz.com/911968/bill-gates-the-robot-that-takes-your-job-should-pay-taxes
- Acemoglu, D. & Restrepo, P. “Automation and New Tasks,” Journal of Economic Perspectives, Vol. 33, 2019, https://pubs.aeaweb.org/doi/pdfplus/10.1257/jep.33.2.3
- The Korea Times, “Korea takes first step to introduce ‘robot tax’,” The Korea Times, August 7, 2017, https://www.koreatimes.co.kr/business/tech-science/20170807/korea-takes-first-step-to-introduce-robot-tax
- WARC, “South Korea’s ‘Robot Tax’,” WARC, 2019, https://www.warc.com/newsandopinion/news/south-koreas-robot-tax/en-gb/39118
- [Reference removed – unverified data]
- Gardner, M. “Amazon’s Federal Tax Avoidance 2017-2019,” Institute on Taxation and Economic Policy, 2020, https://itep.org/amazon-tax-avoidance-2017-2019
- The Verge, “Meta releases the biggest and best open-source AI model yet,” The Verge, July 23, 2024, https://www.theverge.com/2024/7/23/24204055/meta-ai-llama-3-1-open-source-assistant-openai-chatgpt
- European Commission, “Artificial Intelligence Act,” EUR-Lex, 2024, https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:32024R1689
- e-Estonia, “Digitising taxation secures Estonia’s #1 position,” e-Estonia, 2024, https://e-estonia.com/digitising-taxation-secures-estonias-nr-1-position-in-tax-competitiveness-index/
- Decidim, “Open source digital infrastructure for participatory democracy,” Decidim.org, 2024, https://decidim.org/
- Writers Guild of America, “2023 Strike Settlement: AI Protections Summary,” WGA, 2023, https://www.wga.org/contracts/contracts/mba-2023/ai-protections
- Supply Chain Dive, “ILWU, PMA reach tentative deal on key issues,” Supply Chain Dive, 2023, https://www.supplychaindive.com/news/ILWU-PMA-reach-deal-key-issues/648185/
- IndustriALL Europe, “IG Metall Agreement for Metal Workers 2024,” IndustriALL Europe, 2024, https://www.industriall-europe.eu/documents/upload/2024/12/638700225694851701_IG_Metall_-_ME_agreement_2024-2.pdf
- Penn Carey Law, “The German Codetermination Act of 1976,” Journal of Comparative Corporate Law, 1979, https://scholarship.law.upenn.edu/context/jil/article/1023/viewcontent/MertensSchanze2J.Comp.Corp.L._Sec.Reg.75_281979_29.pdf
- [Reference removed – example unverified]
- General reference to Uber’s documented multi-billion dollar losses during market capture phase
- BroadbandNow, “Open Access Broadband Networks in the United States,” BroadbandNow Research, 2024, https://broadbandnow.com/research/open-access-networks
- European Parliament, “Robots and artificial intelligence: MEPs call for EU-wide liability rules,” European Parliament Press Release, 2017, https://www.europarl.europa.eu/news/en/press-room/20170210IPR61808/robots-and-artificial-intelligence-meps-call-for-eu-wide-liability-rules
- Scholz, T. “Platform Cooperatives: Solving the Capital Problem,” Platform Cooperativism Consortium, 2024, https://platform.coop/research/capital-solutions-2024