Nobel Prize-winning economist Daron Acemoglu has issued a warning that multiple factors are steering artificial intelligence toward a "labor-displacing" model of societal transformation. He cautions that the so-called "AI disruption of everything" could ultimately replace human workers, potentially leading to severe and irreversible consequences for social order. This latest perspective from the laureate offers a comprehensive response to the "2028 AI Doomsday Prophecy" recently published by Citrini Research, a report that triggered panic selling in global financial markets. Citrini's report outlines a dystopian vision of an AI-shaped future, predicting that despite an unexpected surge in global AI productivity by 2028, the complete upheaval of white-collar employment could trigger a "global economic plague."

While headlines, industry hype, and employers suggest an imminent AI super-revolution that will quickly make workers more efficient and successful, Acemoglu provided a more measured and unsettlingly pessimistic view in a recent media interview. He acknowledged that recent progress, particularly in Agentic AI focused on "AI-agent全能 workflows," has advanced faster than he personally anticipated. However, concerning reliability, reasoning ability, and understanding of the real world, he believes current AI systems remain deficient. This implies that any scenario involving widespread, immediate, and breakthrough improvements in corporate productivity is unlikely to materialize in the short term.

Yet, the uncertainty about the future direction is far greater than during any previous technological shift, Acemoglu warned. He highlighted that tech giants are overwhelmingly prioritizing the large-scale replacement of workers with AI rather than fostering positive complementarity with the human workforce. According to Acemoglu, this approach could create a false boom in productivity gains resembling a "Phantom GDP" or lead to serious societal repercussions. He argues that the greatest economic benefits would come from "worker-friendly AI"—a development paradigm that enhances human capabilities, enabling workers to perform more complex, higher-value tasks. However, he stated that current business incentives, market structures, and policy frameworks are skewed towards labor substitution.

If the direction is not altered, he warns, mass job displacement—particularly among the global white-collar workforce—could place unprecedented pressure on labor markets, significantly depress wages, and undermine the stability of human social institutions. Acemoglu's view that large-scale AI-driven job replacement could spell economic disaster aligns perfectly with the thesis of Citrini Research. Citrini's proposed mechanism for an "AI Boom Crisis" is as follows: AI agents drive the replacement of white-collar jobs, leading to declining wages and consumer purchasing power, and ultimately resulting in "Phantom GDP"—a situation where GDP and productivity metrics continue to grow, but the human consumption engine, which constitutes about 70% of GDP, stalls, creating a "consumption-less false boom."

Under this dystopian mechanism, the long-standing consumption-driven economy (the article notes current high consumption share) is eroded. This leads to negative feedback for risk assets like stocks at elevated levels, potentially even causing unemployment to surge into double digits, culminating in a significant downturn in global equity markets—a "post-hoc" disaster narrative. Citrini Research effectively splits the simplistic "AI equals rising productivity/margins" story into a conflict between "market prosperity" and "real economic weakness."

Acemoglu also emphasized that shaping AI's trajectory requires both public discourse and policy intervention from governments worldwide. He criticized the grand narrative of the "AI race" among tech giants, calling such purely virtual competition misleading and potentially harmful to employment and economic growth. In contrast, he advocates for a broader focus on practical applications in sectors like healthcare and manufacturing that can tangibly improve societal longevity and productivity.

Daron Acemoglu, who won the Nobel Prize in Economics in 2024 alongside Simon Johnson and James A. Robinson for his work on "how institutions are formed and how they affect economic prosperity," is particularly watched by financial market investors today. His academic background suggests he can analyze how AI, robotics, and automation might reshape future productivity, job structures, and income distribution through the frameworks of macroeconomics, labor, and technology. His comments can potentially influence stock market investment trends.

The pessimistic "AI disruption of everything" narrative has gained significant traction since the beginning of the year. Economists, including Acemoglu, increasingly worry that this disruptive effect could, fueled by elements like high unemployment, false productivity booms, and the "Phantom GDP effect," lead to a collapse of civilization and order. The pessimistic tone since February stems from growing market concerns that viral AI agent workflows, like those from Claude Cowork and OpenClaw, could undermine the entire software empire built on SaaS seat-based subscription models. This triggered rare sell-offs that quickly spread to insurance, real estate, trucking, and any industry reliant on seat-based or labor-intensive business models—sectors perceived as ripe for complete AI disruption.

Not only US stocks but software sectors globally have continued to decline heavily since February under this "AI disruption" panic. Despite a surge in buybacks from US software companies, investors remain unconvinced, as the core fear is the long-term fundamental reshaping of business models by AI agents like Claude Cowork and OpenClaw. The "Anthropic storm" that battered software stocks continues to ripple through global equity markets, with selling accelerating and spreading to wealth advisory, management, real estate consulting, and other traditional sectors seen as vulnerable to AI disruption.

The market's pessimistic "AI disruption" expectation has hit various industry sectors like dominoes, from software, SaaS, and private equity to insurance, traditional investment banking, wealth management, real estate, property management, and even logistics, with stocks "taking turns to plummet." AI has, over the past three to four weeks, seemingly swept through traditional sectors one by one, prompting investors to rapidly sell potential "losers."

Both Acemoglu's latest views and Citrini Research's "2028 AI Doomsday Prophecy" focus on how, under current commercial incentives, capital market preferences, and policy frameworks, the mainstream path of AI technological advancement is increasingly偏向 "labor substitution" rather than "human augmentation." Citrini provides a stress-test doomsday scenario, while Acemoglu sounds an alarm from an institutional economics perspective: if the AI technological path continues to be driven solely by capital return maximization, AI might first bring a series of income and distribution shocks rather than efficiency dividends or productivity prosperity.

The core reason Citrini Research's report, dubbed the "2028 AI Doomsday Prophecy," quickly ignited market volatility lies not in presenting new facts, but in offering a structurally complete, tradable left-tail scenario. Written as a "macro memo looking back from June 2028," it poses a counterintuitive proposition: "What if the continued validation of the AI bull narrative turns out to be bearish for the economy and markets?" While AI could indeed create a "partially dystopian" economic shock, a "global economic plague" is not the baseline scenario but a left-tail risk requiring multiple conditions to spiral out of control simultaneously.

An IMF forecast indicates that around 60% of jobs in advanced economies could be affected by AI, with roughly half potentially experiencing productivity enhancement and the other half closer to reduced labor demand and productivity stagnation. The World Economic Forum suggests that 40% of employers expect to cut staff in tasks automatable by AI, but by 2030, AI and information processing technologies combined could still create 110 million jobs while displacing 90 million. Recent labor research from Anthropic also found that currently the most exposed professions indeed include programmers, customer service representatives, and financial analysts, but no significant rise in unemployment within these fields has been observed so far, only slight slowing in hiring for individuals aged 22 to 25.

In other words, the most immediate risk is not "mass unemployment tomorrow," but the initial erosion of white-collar, entry-level, and high-education/high-wage service sector roles, with effects then transmitting to the macroeconomic level through wages, consumption, and social mobility. Economists, including Acemoglu, repeatedly emphasize that AI's trajectory can be shaped. Government procurement, novel tax system designs for the AI era, competition policy, intellectual property arrangements, vocational training, and redistribution mechanisms will all determine whether AI becomes an "amplifier of expert capability" or a "wage suppression machine." Acemoglu states that whether AI evolves into an economic plague depends on whether we build it as a "system for replacing people" or a "system for empowering humans."