Hundreds of economists warn: AI will trigger more drastic economic changes than the Industrial Revolution, and immediate action must be taken.

Prelude to a Structural Transformation

In March 2025, a public letter consisting of just four sentences was issued by the Stanford University Digital Economy Lab, rapidly gathering signatures from over 200 economists, computer scientists, and tech executives, including 16 Nobel laureates and AI pioneer Yoshua Bengio. This letter, concise to the point of restraint, was in fact one of the most severe warnings to global policymakers: the economic impact of AI is no longer a futurist’s speculation, but an imminent structural earthquake.

The letter states: “In the next decade, AI could become far more powerful than it is today. This will drive an economic transformation even greater than the Industrial Revolution, but in a much shorter timeframe.” This statement implicitly highlights a key global trend: the acceleration of technological change is exceeding human society's capacity to adapt. The Industrial Revolution spanned about a century, while an AI-driven transformation may achieve its initial reshaping within a decade. This is not merely a matter of speed but of governance—traditional labor market adjustment mechanisms, social safety nets, and retraining programs are almost certain to fail under this compressed timeline.

Why Now?

The timing of this open letter is by no means coincidental. Since the advent of large language models like GPT-4, the capability curve of AI has become exceptionally steep. From code generation to legal documents, from medical diagnosis to financial analysis, AI is rapidly penetrating high-skilled white-collar jobs, a stark departure from past automation that primarily impacted blue-collar roles. Economists have come to realize that when creative work, decision analysis, and professional services also begin to be replaced by AI, the foundation of the social contract—the linear path of “education → employment → income → social status”—may collapse.

In a separate statement, Bengio emphasized: “We must make intentional, collective, democratic choices, rather than letting market forces develop on their own, risking leaving the majority of citizens behind.” This captures the core issue: technological change itself does not necessarily bring social progress; it is merely a force that requires institutional guidance. Current market incentives are almost entirely skewed toward maximizing efficiency—that is, accomplishing more tasks with less labor. Under this logic, mass unemployment is not a bug, but a feature.

Three Dimensions of Structural Impact

First, the polarization of the labor market. High-skilled positions are no longer safe, and mid-to-low-skilled jobs are equally under pressure. AI may simultaneously replace programmers and customer service representatives, accountants and assembly line workers. This means the traditional promise of “upward mobility through education” becomes fragile—even acquiring cutting-edge skills may lead to obsolescence within a few years due to AI iteration. The risk of social class stratification intensifies.

Second, the imbalance in the distribution of capital and labor widens further. If AI greatly increases the return on capital (since AI is essentially a capital-intensive technology), while labor’s share of income continues to decline, inequality will sharply worsen. Every technological revolution in economic history has been accompanied by a period of rising inequality, and the intensity of the AI transformation may make previous adjustment periods seem mild.Third, a vacuum in global governance. Policy tools at the national level (such as unemployment insurance, minimum wage, tax redistribution) have struggled to cope with the business models of multinational tech companies and the global deployment of AI. This letter is not only an appeal to the U.S. government but also a collective call to international organizations, the G20, and regulatory agencies around the world. However, there is currently a lack of an effective global platform to coordinate AI economic governance—UN reform is sluggish, the WTO has stalled on digital trade rules, and countries' AI regulatory paths are going their own separate ways (the EU's AI Act emphasizes risk classification, China focuses on the balance between development and security, while the U.S. leans more toward industry voluntary commitments). There is a huge institutional gap between fragmented governance and the unified impact of AI.

Long-term Trends: From "Dealing with Unemployment" to "Redefining Work"

The term "complements humans" used by economists is thought-provoking. It implies a long-term strategic goal: not to prevent AI from replacing humans, but to design institutions that make AI a tool to enhance human capabilities rather than a substitute. This requires rethinking the education system (shifting from knowledge transmission to fostering critical thinking and creativity), the tax system (discussions on taxing robots or AI may move from the fringe to the mainstream), and social security (an upgraded version of universal basic income or employment insurance).

The core proposition in the letter is to establish "incentives, guardrails, and institutions." Incentives mean adjusting market signals, such as using tax breaks to encourage companies to invest in human-augmenting AI rather than purely replacement AI; guardrails mean setting red lines, such as prohibiting fully automated decision-making in certain high-risk areas; institutions mean establishing or strengthening specialized regulatory agencies—just as the Industrial Revolution gave rise to labor laws and unions, the AI revolution also requires a new social contract.

Global Differences in Perspective

It is worth noting that the signatories are mainly scholars and entrepreneurs from the English-speaking world. This reflects that AI technology R&D is concentrated in Silicon Valley, Boston, London, and Beijing, but the economic impact will be global. Developing countries may face higher risks: their social security systems are weaker, their labor structures are more fragile, and AI deployment often prioritizes places with the highest cost-effectiveness (usually manufacturing and service outsourcing centers with low labor costs). For example, India's call centers and the Philippines' business process outsourcing industry are already feeling the pressure. Although this open letter originates from top U.S. universities, its appeal should be taken seriously by Global South countries, because an AI transformation lacking redistribution mechanisms could further widen the North-South divide.

Conclusion: The Window of Opportunity Is ClosingThe most unsettling aspect of this letter is its sense of urgency. Economists are typically cautious, rarely making assertions such as "we must act now." Yet when hundreds of leading scholars unanimously agree that "we must act now," it transcends academic debate and becomes a global policy signal. The Industrial Revolution took a century to shape the economic structure we know; the AI transformation may take only a decade. In this decade, society’s adaptability, innovation capacity, and governance wisdom will face unprecedented tests.

The key issue is not whether change will happen, but how it will unfold—whether chaotically, fraught with conflict, or orderly and inclusive. This open letter poses the questions, but the answers lie in the hands of global policymakers. And time is not on their side.