THE MISATTRIBUTION EFFECT
How Wealth Concentration and Geopolitical Shocks
Are Being Mistaken for the Impact of Artificial Intelligence
Research Outline & High-Level Summary
Focused on the Economies of the United States and the United Kingdom
DRAFT — April 2026
Tracy Work (with AI)
Executive Summary
The prevailing narrative in public discourse, media coverage, and policy circles holds that artificial intelligence is the dominant force reshaping employment, wages, and economic opportunity in the United States and the United Kingdom. This article challenges that narrative with data.
The central thesis is straightforward: much of what the public perceives as the economic impact of AI is, in fact, the continuation of a decades-long concentration of wealth at the top of the income distribution, compounded by severe geopolitical shocks—most notably the Russia-Ukraine war beginning in 2022 and the Israel/US-Iran conflict erupting in early 2026—that have driven inflation, eroded real household incomes, and destabilized energy markets in ways that have nothing to do with artificial intelligence.
The data bears this out across multiple dimensions:
• Federal Reserve data shows the top 1% of U.S. households held 31.7% of all wealth by Q3 2025—the highest share since tracking began in 1989—a trend that predates generative AI by decades.
• UK wealth inequality shows a similar pattern: the richest 10% hold 43% of all wealth while the bottom 50% own just 9%, driven primarily by property wealth concentration and intergenerational transfers, not technology adoption.
• Confirmed AI-related job losses in the U.S. through 2025 total approximately 55,000—a fraction of normal labor market turnover—while AI-related job creation in 2024 alone reached roughly 119,900 positions.
• Brookings Institution research published in late 2025 found “no AI jobs apocalypse” and concluded that occupational changes since ChatGPT’s launch are “strikingly consistent with past precedent” for technology adoption.
• The Ukraine war drove UK inflation to a 41-year high of 11.1% in October 2022. The 2026 Iran conflict has pushed oil above $120/barrel with cascading effects on consumer prices—economic shocks wholly unrelated to AI.
This article argues that conflating these distinct economic forces is not merely an academic error—it risks misdirecting policy, inflaming unnecessary public anxiety, and distracting from the real structural challenges facing workers in both countries. Drawing on the author’s 2018 white paper presented before Congressional committees on AI’s anticipated employment impact, this piece demonstrates that AI adoption is, in fact, following a trajectory remarkably consistent with historical patterns of major technology integration.
Part I: The Wealth Concentration Story
Chapter 1: The Numbers Behind the Narrative
This chapter establishes the foundational economic data demonstrating that wealth and income inequality in both the U.S. and UK have been accelerating along trajectories that began well before the emergence of generative AI.
United States
• The top 1% of U.S. households held approximately 30.5% of wealth in Q1 2024, rising to 31.7% by Q3 2025—the highest recorded level since the Federal Reserve began tracking in 1989. The bottom 50% held just 2.5%.
• From 1989 to 2019, wealth concentration at the top 1% and top 10% increased steadily, driven largely by corporate stock ownership concentration. The bottom 50% experienced effectively zero net wealth growth over this 30-year period.
• During the COVID-19 pandemic, U.S. billionaire wealth increased by 70%, with 2020 marking the steepest increase in billionaires’ share on record.
• The top 10% of income earners accounted for nearly half of all consumer spending by Q2 2025. Stock market gains—particularly in AI-related equities—disproportionately benefit the wealthy, since 87% of stock-owning Americans earn $100,000+.
• Bank of America data from December 2025 shows wage growth of 3% for higher-income households versus just 1.1% for low-income households.
United Kingdom
• The UK has the 9th most unequal income distribution among 38 OECD countries. The top fifth take 36% of income and 63% of wealth; the bottom fifth have 8% of income and just 0.5% of wealth.
• By 2023, the richest 50 families held more wealth than the bottom half of the UK population (34.1 million people). At current growth rates, the wealth of the richest 200 families will exceed total UK GDP by 2035.
• Wealth inequality actually increased during COVID-19, driven by enforced savings among the affluent and unprecedented asset price rises—an unusual pattern for an economic downturn.
• Younger generations are increasingly locked out of wealth accumulation: only 36% of those born in the 1980s owned homes by age 30, compared to over 60% of those born in the 1950s-60s.
• Median household incomes fell in real terms in 2023/24, with the lowest 10% of incomes falling 7%—despite lower inflation and a stronger labor market.
Key argument: These trends have been underway for three to four decades. The acceleration in inequality observable since 2020 is driven by pandemic-era asset inflation, stock market concentration, and housing dynamics—not by AI deployment, which was still nascent during most of this period.
Chapter 2: The Wealth Transfer Gary Vaynerchuk Describes
This chapter draws on Gary Vaynerchuk’s argument about wealth concentration and economic disruption, contextualizing it within the broader data. Vaynerchuk’s thesis—that we are witnessing a generational wealth transfer and economic restructuring driven by attention economics and platform dynamics, not AI per se—aligns with the macro data.
• Vaynerchuk argues that what many perceive as AI disruption is really the latest iteration of a pattern that has repeated with every major technology: calculators, the internet, the tractor. The creative and strategic thrive; the commodity-skill worker faces displacement.
• This framing is supported by the data: the K-shaped recovery post-COVID, where asset-rich households surged ahead while wage-dependent households stagnated, occurred before generative AI entered mainstream use.
• The “shift of wealth” Vaynerchuk describes is measurable: the gap between asset-based wealth (stocks, property) and wage-based wealth has widened at an accelerating rate since 2020.
Part II: Geopolitical Shocks as the Real Economic Disruptors
Chapter 3: The Ukraine War and the Inflation Shock
This chapter documents the enormous economic damage inflicted by the Russia-Ukraine conflict on the U.S. and UK economies—damage that has been widely felt by ordinary workers and households but is increasingly, and incorrectly, attributed to AI.
• The UK economy was hit hardest through energy prices. UK inflation peaked at 11.1% in October 2022, a 41-year high. Energy bills rose 54% in April 2022 under the new price cap.
• Research using computable general equilibrium modeling found that the US, UK, and EU experienced GDP declines of 0.5–3%, household income drops of 2–4%, and consumption decreases of 1.5–3.5% as direct consequences of war-driven energy price spikes.
• The IMF noted that more than half of advanced economies already had inflation above 5% before the invasion; the war made an already difficult situation dramatically worse.
• The OBR estimated the invasion reduced UK GDP and eroded real household incomes primarily through higher fuel and energy prices that cascaded into the cost of all goods and services.
• Consumer confidence in the UK fell to its lowest since January 2021, well before generative AI entered mainstream conversation.
Chapter 4: The 2026 Iran Conflict — A Second Energy Shock
This chapter covers the February-March 2026 Israel/US-Iran conflict and its immediate economic consequences, providing the most current evidence of geopolitical forces—not AI—driving economic anxiety.
• The conflict disrupted approximately 20 million barrels per day of oil transit through the Strait of Hormuz—roughly 20% of global petroleum supply.
• Brent crude surged 55% from $72.48 to $112.57 between late February and late March 2026. Some analysts project prices could reach $150.
• U.S. gasoline prices rose above $4/gallon, the highest since late 2023. UK petrol prices rose sharply, and household gas bills are expected to increase later in 2026.
• Fertilizer prices jumped 40%, threatening global food costs. The Bank of England’s anticipated interest rate cuts are now unlikely, and hikes are possible.
• The House of Commons Library concluded that the conflict’s energy price disruption is expected to lead directly to higher UK inflation and lower GDP growth.
Key argument: These are massive, tangible, immediate economic shocks affecting every household in both countries. They have nothing whatsoever to do with artificial intelligence. Yet public discourse increasingly attributes the resulting economic pain to “AI disruption.”
Part III: What the AI Data Actually Shows
Chapter 5: AI Job Displacement — The Gap Between Headlines and Reality
This chapter presents the actual data on AI’s impact on employment, which tells a far more nuanced story than the apocalyptic headlines suggest.
• Confirmed AI-related job losses in the U.S. through 2025 total approximately 55,000, with over 75% occurring after 2023. This represents a tiny fraction of overall labor market turnover.
• Meanwhile, AI-related job creation in 2024 alone reached approximately 119,900 direct positions—more than double the cumulative AI job losses—including 8,900+ in AI development and over 110,000 in data center construction.
• The Information Technology and Innovation Foundation concluded that “the employment gains from AI and the data center buildout dwarf the displacement effects from automation.”
• The WEF’s Future of Jobs Report 2025 projects 92 million roles displaced by 2030 alongside 170 million new roles—a net gain of 78 million jobs globally.
• Critically, Brookings Institution research (October 2025) found that occupational changes since ChatGPT’s launch are “strikingly consistent with past precedent” and that changes in the occupational mix “predate ChatGPT’s launch, suggesting AI may not be the primary driver.”
• The same Brookings study found “no pattern of increasing AI exposure among the unemployed.”
Chapter 6: AI Adoption Is Following Historical Patterns
This chapter connects back to the author’s 2018 white paper and demonstrates that AI’s integration into the economy is following the same trajectory as prior transformative technologies.
• The 2018 white paper argued that, historically, major technological advances generate more jobs on the whole, and that AI was likely to be a net creator of jobs in sectors with elastic demand. This prediction has held.
• The paper warned that disruption would be concentrated in areas where AI serves as a direct labor substitute (e.g., trucking, routine tasks) and among low-income, low-education workers. This is exactly what is emerging.
• St. Louis Federal Reserve data shows generative AI adoption at 54.6% of U.S. adults by August 2025—outpacing both the PC (19.7% three years after IBM PC) and the internet (30.1% three years after commercial launch).
• However, adoption speed does not equal economic disruption speed. Transformative technologies like the computer and internet took decades for their full labor market impacts to materialize, because adoption requires complementary investments, cultural shifts, and regulation.
• Most AI users (95%+) are on free-tier models, suggesting deep integration into work processes remains early-stage.
• Microsoft’s 2025 AI Diffusion Report identifies AI as the “fastest-spreading technology in human history” but notes that adoption tracks existing economic infrastructure—fastest where electricity, connectivity, and computing foundations already exist.
Key argument: AI is following the same adoption S-curve as previous general-purpose technologies, just at a faster rate. But faster adoption does not mean faster economic destruction. The evidence shows AI is creating more jobs than it is displacing, exactly as the 2018 white paper predicted for technologies adopted in sectors with elastic demand.
Part IV: The Misattribution Problem
Chapter 7: Why We Blame AI — Cognitive and Media Dynamics
This chapter explores why the misattribution occurs, drawing on media incentives, cognitive biases, and the political economy of technology narratives.
• AI is a visible, tangible, and narratively compelling explanation for economic anxiety. Wealth concentration and geopolitical energy markets are abstract and diffuse.
• Media incentives favor AI-as-villain stories: they generate engagement, clicks, and fear. “Wealth inequality continues its 40-year trend” is a less compelling headline than “AI is coming for your job.”
• The technology industry itself contributes to the narrative: overstating AI’s transformative power serves investment valuations and creates demand for AI consulting and products.
• Availability bias means that because people interact with AI tools daily, they over-attribute economic changes to AI while under-weighting less visible forces like interest rate policy, energy markets, and wealth concentration.
• Survey data illustrates the gap between perception and reality: 51% of American workers worry about AI replacing their jobs by 2026, yet confirmed AI job losses represent less than 0.04% of total U.S. employment.
Chapter 8: Policy Implications — Solving the Right Problems
This chapter argues that misattributing economic pain to AI leads to misguided policy responses, and outlines what data-driven policy should focus on instead.
• If policymakers focus exclusively on AI regulation and retraining, they miss the structural forces actually driving inequality: tax policy, housing costs, wage stagnation, and geopolitical energy dependence.
• The UK and U.S. need policies addressing wealth concentration directly: progressive taxation, housing affordability, and support for wage growth at the bottom of the distribution.
• Energy policy and geopolitical resilience are urgent priorities: diversification away from Middle East energy dependence, strategic reserves, and renewable energy investment.
• AI policy remains important but should be proportionate: the 2018 white paper’s recommendations for targeted sector-specific transition assistance, retraining, and a pro-innovation environment remain sound.
• The greatest risk is that AI becomes a scapegoat that allows the actual drivers of economic inequality to go unaddressed.
Conclusion: Seeing Clearly
The evidence presented in this article points to a clear conclusion: the economic anxiety felt by millions of workers and families in the United States and the United Kingdom is real, but its causes are being systematically misidentified.
The data shows that:
1. Wealth concentration has been accelerating for four decades, driven by asset price dynamics, stock market concentration, housing markets, and policy choices—not by AI.
2. Two major geopolitical shocks—the Ukraine war and the Iran conflict—have delivered severe blows to household budgets through energy prices, inflation, and interest rate responses.
3. AI’s actual impact on employment, while real and deserving of thoughtful policy, remains modest by any measure: net job creation exceeds displacement, and the pace of occupational change is consistent with historical technology adoption.
4. The 2018 white paper’s core predictions have largely held: AI is a net job creator in elastic-demand sectors, disruption is concentrated among routine-task workers, and the technology is integrating into the economy along familiar historical patterns.
Conflating these distinct forces is not just analytically wrong—it is dangerous. It directs public anxiety toward the wrong target, promotes policy responses that miss the mark, and distracts from the urgent structural challenges that are actually eroding economic opportunity for working people on both sides of the Atlantic.
The challenge before us is not to stop or slow AI—it is to honestly diagnose why so many people are struggling, and to have the courage to address the real causes: a wealth distribution system that has been concentrating gains at the top for a generation, and a geopolitical environment that continues to deliver energy and price shocks to economies that remain dangerously exposed.
Key Data Sources & References
The following represents a preliminary list of key data sources identified during initial research. A comprehensive bibliography will accompany the full article.
Wealth & Inequality Data
• Federal Reserve Distributional Financial Accounts (quarterly, 1989–present)
• UK Office for National Statistics, Wealth and Assets Survey (2006–2022)
• World Inequality Database (wid.world)
• House of Commons Library, Income Inequality in the UK (updated April 2025)
• Institute for Fiscal Studies, Deaton Review of Inequalities
• CBS News / Moody’s Analytics analysis of Federal Reserve wealth data (January 2026)
AI Employment Impact
• Brookings Institution, “New Data Show No AI Jobs Apocalypse—For Now” (October 2025)
• St. Louis Federal Reserve, “The State of Generative AI Adoption in 2025” (November 2025)
• Information Technology and Innovation Foundation, “AI’s Job Impact: Gains Outpace Losses” (December 2025)
• World Economic Forum, Future of Jobs Report 2025
• IMF assessment of global AI job exposure (2024)
• St. Louis Federal Reserve, “Is AI Contributing to Rising Unemployment?” (August 2025)
Geopolitical & Energy Impact
• UK Office for Budget Responsibility, analysis of Ukraine invasion impact (March 2022)
• House of Commons Library, “Middle East Conflict and the UK Economy” (April 2026)
• Deloitte Insights, “Iran and Middle East Conflict Impacts Global Economy” (March 2026)
• World Economic Forum, “The Global Price Tag of War in the Middle East” (March 2026)
• IMF, “The Long-Lasting Economic Shock of War”
• Oxford Economics, impact scenarios for Iran-Israel escalation (June 2025)
Technology Adoption Patterns
• Microsoft AI Diffusion Report 2025
• Epoch AI, “After the ChatGPT Moment: Measuring AI’s Adoption” (July 2025)
• St. Louis Federal Reserve, “The Rapid Adoption of Generative AI” (October 2024)
• [Author], “Meeting the AI and Employment Challenge” — Intel Corporation white paper (2018)