By Jim Shimabukuro (assisted by ChatGPT)
Editor
Introduction: Elon Musk predicted, at the US-Saudi Forum 19 Nov 2025, that “work will be optional” in approximately 10-to-20 years as a result of advances in AI technology. I asked ChatGPT to search the current (2025) literature for (1) the three strongest arguments FOR Musk’s prediction and (2) the three strongest arguments AGAINST his prediction. I added that the arguments need not refer to Musk or the US-Saudi forum. -js
1. Rapid and large productivity gains from enterprise AI could create a post-labor surplus (McKinsey, 2025)
Source: McKinsey & Company, AI in the workplace: A report for 2025 (prompted by Reid Hoffman’s Superagency), January 28, 2025. https://www.mckinsey.com/capabilities/tech-and-ai/our-insights/superagency-in-the-workplace-empowering-people-to-unlock-ais-full-potential-at-work. (McKinsey & Company)
This McKinsey report is an industry-scale assessment of generative AI adoption and its corporate economic potential. The paper synthesizes firm-level case studies and macro projections to estimate that AI offers very large productivity upside: McKinsey’s internal research, cited in the report, sizes the long-term corporate opportunity in the trillions of dollars and treats the current era of LLMs as a new “cognitive industrial revolution.”
The argument in favor of the “work optional” thesis follows this logic: if AI (and robotics where relevant) can sustain a structural, multi-trillion-dollar expansion of output per unit of human labor, then the economic constraint that requires most people to sell labor for subsistence weakens or disappears. Once capital and software capture vastly more surplus value with little or no labor input, societies could redistribute that surplus (via different fiscal arrangements, UBI-style transfers (Universal Basic Income), sovereign funds, corporate profit-sharing, etc.), enabling many people to stop doing paid work without falling into poverty.
The relevance is practical and existential: policymakers, firms, and social planners must decide whether to adjust tax, redistribution and labor institutions toward a future where basic consumption is decoupled from paid employment.
Direct quote: “Artificial intelligence has arrived in the workplace and has the potential to be as transformative as the steam engine was to the 19th-century Industrial Revolution.” (McKinsey & Company)
2. Measurable, rapid productivity improvements from generative AI imply much less work time may be required (Penn Wharton Budget Model, Sept 8, 2025)
Source: Penn Wharton Budget Model, The Projected Impact of Generative AI on Future Productivity Growth, September 8, 2025. https://budgetmodel.wharton.upenn.edu/issues/2025/9/8/projected-impact-of-generative-ai-on-future-productivity-growth. (Penn Wharton Budget Model)
This policy-oriented economic modeling brief applies a task-based framework (building on Acemoglu and others) to project productivity and GDP impacts from generative AI. Its sober model estimates that generative AI raises overall productivity and GDP by nontrivial amounts (they report a 1.5% increase in level by 2035 and larger gains longer term). The core supportive argument is structural: even moderate, persistent gains in total factor productivity (TFP) compound over time into large increases in output per worker.
If machines consistently produce more of what society values with less human work input — and if those gains can be captured as income (capital returns, corporate profits, or public revenue) — then society can afford to meet people’s needs with less paid labor. The model’s policy implication is twofold: first, AI can materially reduce the labor hours required to sustain current living standards; second, whether that leads to voluntary work depends on institutional choices about distribution (taxes, transfers, labor law).
The Penn-Wharton results matter because they move the discussion from speculative utopia to quantitative feasibility: measured productivity effects are already visible and, if sustained, point to a plausible economic pathway where work becomes a choice rather than a necessity.
Direct quote (one sentence): “We estimate that AI will increase productivity and GDP by 1.5% by 2035, nearly 3% by 2055, and 3.7% by 2075.” (Penn Wharton Budget Model)
3. If AI-generated surplus is redistributed (UBI/AI dividends), wage work can become optional (Alexandre Rigaud, ResearchGate paper, Sep 1, 2025)
Source: Alexandre Rigaud, Universal Basic Income in the Age of Artificial Intelligence: Redistributing AI-Generated Profits for a Sustainable Future, ResearchGate (public upload), September 1, 2025. https://www.researchgate.net/publication/395135368_Universal_Basic_Income_in_the_Age_of_Artificial_Intelligence_Redistributing_AI-Generated_Profits_for_a_Sustainable_Future. (ResearchGate)
Rigaud’s paper is explicitly normative and policy-technical: it accepts that AI will produce large surplus profits by displacing labor across cognitive and manual tasks, and it develops an implementable blueprint for capturing that surplus to fund a universal basic income (UBI) financed by an “AI dividend” or automation tax. The argument supporting Musk’s public claim is institutional: the technological possibility of rendering paid work unnecessary is distinct from the political choice to make it socially feasible.
Rigaud demonstrates that, under plausible levels of automation-driven profit, governments could finance a meaningful UBI without catastrophic fiscal stress if they adopt redistribution mechanisms targeted at AI returns (taxes on AI deployment, profit-sharing, sovereign AI funds).
This matters because it converts a speculative technological claim (“machines can do most work”) into a pragmatic social program: if redistributive institutions keep pace, people could reduce or end paid work and live off social income, making work largely optional. The paper also traces precedents (resource sovereign funds, Alaska’s Permanent Fund) to show the political economy is tractable.
Direct quote (one sentence): “If AI amplifies profits, then there is a moral, economic, and political imperative to use these gains to fund a universal basic income.” (ResearchGate)
Three strongest arguments REJECTING Musk’s prediction
1. Uneven adoption, rising wealth concentration, and policy limits make a “work optional” society unlikely without massive redistribution (IMF working papers, 2025)
Source: IMF (multiple 2025 working papers summarized), notably AI Adoption and Inequality, WP/25/68 and The Global Impact of AI: Mind the Gap, WP/25/76 (2025). (Examples and summaries available at IMF eLibrary and working paper listings.) https://www.elibrary.imf.org/view/journals/001/2025/068/article-A001-en.pdf. (elibrary.imf.org)
IMF research in 2025 stresses that AI’s economic benefits are neither automatic nor evenly distributed. The rejecting argument runs like this: even if AI raises aggregate productivity, those gains will be captured disproportionately by capital owners, large platforms, and technologically leading countries and firms. Without deliberate, effective redistribution — which the IMF warns faces political, administrative, and international coordination barriers — productivity gains will increase wealth inequality rather than create a broadly shared surplus that enables universal leisure.
Furthermore, IMF authors emphasize that adoption is path-dependent and constrained by institutional capacity, infrastructure, and workforce skills: many emerging and low-income countries lag in AI adoption, and within advanced economies, the winners (high-skill workers and capital owners) reap most gains.
Practically, this means that a world where everyone can stop working rests on politically difficult reforms (taxation on capital, global coordination, strong social safety nets) that are not inevitable. The IMF’s policy framing therefore rejects the idea that technology alone will make work optional; institutional and political change would be required and is far from assured.
Direct quote (one sentence): “We find that while AI may reduce wage inequality by displacing high-income workers, it is likely to substantially increase wealth inequality as these same gains accrue to capital and incumbent firms.” (elibrary.imf.org)
2. Historical and task-based evidence suggests technology creates new work and shifts tasks rather than eliminating the need to work (OECD Employment Outlook 2025)
Source: OECD, OECD Employment Outlook 2025, July 2025. https://www.oecd.org/content/dam/oecd/en/publications/reports/2025/07/oecd-employment-outlook-2025_5345f034/194a947b-en.pdf. (OECD)
The OECD report presents a balanced, evidence-based caution: while generative AI is transformative, history suggests that technological revolutions shift the composition of work rather than make work optional. The OECD frames the current moment as one of competing forces: AI can raise productivity and create new industries and tasks, but it also risks displacing existing jobs and increasing inequality if skill mismatches and regulatory failures aren’t addressed.
The central rejecting argument is empirical and institutional: measured productivity gains so far are concentrated in specific sectors and tasks, new categories of jobs emerge (oversight, data curation, human-AI collaboration), and demographic forces (ageing populations in advanced economies) create labor shortages that actually keep demand for labor high.
The OECD thus warns that the realistic outcome is not mass leisure but transformation in what work looks like, with policy choices (education, reskilling, labor market regulation) shaping whether transitions are benign. This matters because it reframes the debate: policy should focus on managing transitions and upgrading skills, not on assuming a post-work world will happen by technological determinism alone.
Direct quote (one sentence): “Rapid advances in generative AI are fuelling both optimism for renewed productivity growth and concern about job displacement and widening inequalities.” (OECD)
3. Empirical studies show displacement risks and insufficient mitigation; human-AI complementarity and fragmented policy responses make “work optional” unlikely soon (Elsevier / ScienceDirect review article, 2025)
Source: M. Nigar et al., Artificial intelligence and technological unemployment: Understanding trends, technology’s adverse roles, and current mitigation guidelines, ScienceDirect (Elsevier), 2025. https://www.sciencedirect.com/science/article/pii/S2199853125001428. (ScienceDirect)
This 2025 academic review aggregates evidence across sectors (healthcare, education, creative industries) and finds substantial heterogeneity in AI’s labor impacts. The rejecting argument grounded in this literature is methodological and pragmatic: current studies show that AI causes task automation, skill polarization, and displaces certain cohorts (entry-level white-collar, routine cognitive tasks), but policy responses remain fragmented and insufficient to handle structural displacement.
The review concludes that without coherent, large-scale mitigation (comprehensive retraining, income supports, institutional redesign), displacement will increase precarity rather than free people to live without paid work. In short, the technology’s capacity to replace tasks is real, but the social, legal, and institutional infrastructure required to convert that capacity into a widespread option to not work is not present and is unlikely to be built decisively within 10–20 years. This evidence-first skepticism undermines claims that work will become optional soon: technological possibility alone does not equal social reality.
Direct quote (one sentence): “While a growing body of policy responses encourages human-AI complementarity, current measures remain fragmented and insufficient to address the structural risks of workforce displacement.” (ScienceDirect)
Methodological note
ChatGPT focused on 2025 publications (institutional reports and peer-reviewed or policy papers) that are influential in policy and economic debates: McKinsey (industry modeling of corporate productivity potential), Penn-Wharton (formal productivity projections), ResearchGate policy paper on UBI (explicit institutional pathway), IMF working papers (distributional risks and adoption gaps), OECD Employment Outlook (balanced, cross-country evidence), and a 2025 Elsevier literature review (empirical synthesis). All six were chosen because they capture the strongest, most defensible forms of the “for” and “against” arguments in current public and academic discourse.
[End]
Filed under: Uncategorized |
Educational Technology and Change Journal 
Leave a comment