AI's $650B bet risks stalling without worker buy-in, history warns

Four dominant tech companies are on track to spend $650 billion on data centers and infrastructure in 2025 alone, with AI-linked stocks now representing between 50% and 57% of the S&P 500's total value. The prevailing view in Silicon Valley and on Wall Street is that the technology works, its capability is unprecedented, and mass adoption is inevitable. Yet the past several months have revealed mounting friction. In April, a man motivated by anti-AI sentiment attacked OpenAI CEO Sam Altman's home. In May, college graduates entering a job market transformed by AI met commencement speakers—including former Google CEO Eric Schmidt—with boos. In June, Anthropic co-founder Jack Clark publicly urged governments to build regulatory mechanisms to slow AI development, and frontline workers, particularly from Gen Z, began quietly undermining their employers' AI rollouts.

The article argues that these signs reflect a fundamental misunderstanding in corporate leadership: the assumption that society is a passive operating system that will automatically update to accommodate new technologies. In reality, the author contends, technologies scale based on human incentive structures, not abstract mathematical capability. When adoption threatens stakeholder welfare while concentrating all gains at the top, people will build firewalls to freeze the technology in place—a pattern the author terms the "Jersey Pump Principle." In the late 1940s, the automatic shut-off gas nozzle revolutionized fueling technology, yet New Jersey banned self-service gas pumps in 1949 because independent station owners and labor advocates wanted to protect pump attendant jobs. Nearly 80 years later, New Jersey remains a lone holdout, legally locking in thousands of attendant positions.

A contemporary parallel unfolded in San Francisco and Austin with autonomous robotaxis. Tech companies secured regulatory approval to expand driverless fleets 24/7, but the rollout marginalized local drivers, blocked fire department emergency vehicles, and offered residents no upside. Activists discovered that placing cheap orange traffic cones on the vehicles' hoods would trip their sensors into permanent safety stalls, completely immobilizing the cars. Following publicized accidents and intense localized friction, California's DMV suspended Cruise's fleet indefinitely, citing "unreasonable risk to public safety."

History also offers a counter-example: when incentives align, adoption accelerates with virtually zero friction. When Henry Ford rolled out the moving assembly line in 1913, build times for the Model T dropped from 12.5 hours to 93 minutes, but factory-floor monotony caused labor turnover to skyrocket to 370% in 1913, forcing Ford to hire over 50,000 workers just to maintain a baseline of 14,000. In 1914, Ford doubled the industry-standard wage to $5 a day and cut shift length by an hour, offering workers immediate financial security in exchange for accepting assembly-line work. Attrition plummeted, productivity soared, and Ford's profits nearly doubled within two years. Similarly, Toyota, rather than using early robotics as a weapon to cut headcounts, guaranteed factory workers lifetime employment and installed the "Andon Cord"—a physical rope allowing any worker to halt the entire assembly line if they spotted an inefficiency. Because automation was tied to worker empowerment rather than displacement, workers actively partnered with engineers to optimize robots, achieving defect rates lower than American competitors'.

The article contends that current corporate AI deployment follows an extractive model: executives view AI as a tool to reduce headcount, monitor output, and squeeze margins, giving employees no rational incentive to cooperate. Workers facing higher performance quotas or pink slips will quietly slow-roll, sandbag, and sabotage the transition. The author points to the automotive sector's 2023 UAW strike against Ford, GM, and Stellantis as instructive. Autoworkers initially feared that simpler electric drivetrains would mean far fewer assembly jobs—a fear that later research showed was overstated, but perception triggered resistance. In September 2023, the UAW launched its first-ever simultaneous "Stand-Up Strike" against all three companies. The 46-day strike forced carmakers to adopt "just transition" terms: priority transfer rights, elimination of wage tiers, and explicit commitments to bring battery-plant jobs under national union agreements. By removing the threat of human obsolescence, the industry's EV retooling is now moving forward with far less friction—but the author notes the uncomfortable lesson: incentive alignment was forced by the shutdown, not negotiated in advance. For AI executives, the path forward requires structural incentive alignment, ensuring employees see a tangible personal stake in automation—whether relief from systemic burnout, clearer paths to high-value work, or guaranteed career upskilling. Without this alignment, expensive AI technologies will remain underutilized as the collective inertia of workers ensures they do not succeed.