We Are Driving in the Fog: Hundreds of Economists Admit They’re Flying Blind on AI

The conversation around artificial intelligence has shifted from speculative wonder to urgent concern. No longer confined to tech conferences or science fiction, AI’s impact on jobs, wages, and economic stability is now a front-and-center debate in boardrooms, living rooms, and government halls. Yet, despite the growing chorus of warnings, a striking admission has emerged from an unexpected quarter: many of the very experts tasked with forecasting economic trends say they’re essentially guessing when it comes to AI’s long-term effects.

Over 200 economists, including 16 Nobel laureates, recently signed a letter warning that the rapid advancement of AI poses significant risks to employment and economic equity. Their message isn’t that AI will definitely destroy jobs — though that’s a real possibility — but that we lack the tools, data, and models to predict how it will reshape labor markets with any confidence. In essence, we’re driving in heavy fog, and the dashboard lights are flickering.

This uncertainty isn’t due to a lack of effort. Economists have spent decades refining models to forecast the impact of technological change, from the steam engine to the personal computer. But AI is different. It’s not just automating routine tasks; it’s encroaching on cognitive work — writing, analysis, design, even aspects of medical diagnosis and legal reasoning. The speed of adoption, combined with the technology’s ability to learn and improve independently, makes historical analogies unreliable. As one economist put it, we’re trying to predict the weather using only yesterday’s temperature readings.

The letter, which also garnered signatures from prominent figures like Eric Schmidt and Reid Hoffman, calls for immediate action. Not panic, but preparation. The signatories urge governments, businesses, and educational institutions to invest in understanding AI’s economic footprint before it’s too late. That means funding research into labor market transitions, experimenting with policy responses like expanded retraining programs or portable benefits, and creating better real-time data systems to track how AI is actually being used in workplaces — not just how it’s advertised.

Part of the challenge lies in the opacity of AI deployment. Unlike a factory installing robotic arms, where the change is visible and measurable, AI often works behind the scenes. An algorithm might screen job applicants, optimize delivery routes, or generate marketing copy without anyone in the organization fully understanding its inner workings — or its broader economic ripple effects. This “black box” nature makes it harder for economists to trace cause and effect. Did wages stagnate because of AI? Or globalization? Or demographic shifts? Untangling these forces requires granular data that, frankly, we don’t yet collect at scale.

Still, the absence of perfect knowledge shouldn’t paralyze us. History shows that societies don’t need perfect foresight to navigate technological shifts — they need adaptability. During the Industrial Revolution, policymakers didn’t wait for precise models before introducing factory safety laws or public education systems. They acted on observable trends and moral imperatives. Today, we have early signals: productivity gains in certain sectors, job postings declining for roles like copywriters and paralegals, and growing anxiety among knowledge workers about their long-term relevance.

What we do know is that AI’s impact won’t be evenly distributed. Workers with advanced technical skills or those in roles requiring complex human interaction — therapists, teachers, skilled tradespeople — may see their value increase. Meanwhile, those in mid-skill, routine cognitive jobs face the highest risk of displacement. Geographic divides could widen too, as AI hubs concentrate wealth and opportunity in certain cities while other regions lag behind.

This unevenness underscores the need for proactive policy. Not blanket resistance to AI — that would be neither feasible nor wise — but thoughtful stewardship. Ideas like wage insurance, which compensates workers who take lower-paying jobs after displacement, or universal basic income pilots, deserve serious exploration. So too does rethinking education: not just teaching coding, but fostering adaptability, critical thinking, and skills that complement rather than compete with machines.

There’s also a cultural dimension. The economist who shared that living multigenerationally with her parents and kids in Seattle saved everyone money isn’t just talking about housing costs — she’s hinting at a broader truth: resilience often comes from community, not just career. As AI disrupts traditional employment models, we may need to rebuild social safety nets not just around jobs, but around care, learning, and mutual support.

We are, indeed, driving in the fog. But fog doesn’t mean we should stop the car. It means we should slow down, turn on the low beams, and pay close attention to the road ahead — while trusting that we can adjust our course as the mist clears. The economists’ warning isn’t a prediction of doom; it’s a call to humility, preparation, and collective action. The best way to fly blind is to make sure we’re not flying alone.