In December 2025, the Stanford Institute for Economic Policy Research (SIEPR) convened a forum bringing together technologists, economists, business leaders, and government officials to address the economic impacts of artificial intelligence. The meeting revealed a stark divergence in perspectives, with technologists expressing optimism and economists adopting a more cautious outlook.
While past technological advancements have often been met with exuberant predictions from their creators, the current debate surrounding AI is notable for the contrasting views held by those building the technology and those studying its potential effects. Many technologists envision AI as a transformative force, capable of generating substantial economic growth. However, a growing number of economists are skeptical of such claims, suggesting that the actual impact may be far more modest.
MIT Institute Professor Daron Acemoglu, a 2024 Nobel laureate in economic sciences, recently published a paper, “The Simple Macroeconomics of AI,” offering a more conservative estimate of AI’s impact on the U.S. Economy over the next decade. Acemoglu estimates that only approximately 5% of tasks will be profitably performed by AI within that timeframe, leading to a potential GDP boost of around 1%. This contrasts sharply with predictions from organizations like Goldman Sachs, which forecast a $7 trillion increase in global GDP over the next 10 years, and McKinsey, which estimates annual growth between $17.1 and $25.6 trillion. Acemoglu’s analysis suggests these figures are “much less than both the revolutionary changes some are predicting and the less hyperbolic but still substantial improvements” offered by other institutions.
Acemoglu’s research indicates that while nearly 20% of tasks in the U.S. Labor market could be replaced or augmented by AI, the actual implementation of AI will be limited to roughly a quarter of those tasks – the 5% figure – due to the costs often exceeding the benefits. He arrived at this figure by analyzing prior scholarship on tasks exposed to AI and computer vision technologies, combined with expectations around productivity derived from studies of AI deployment in real workplaces. His calculations suggest a total AI-driven productivity increase of approximately 0.7% over the next 10 years.
This divergence in opinion echoes concerns raised in other forums. A recent Forbes article highlighted a similar split, noting that while some economists believe jobs lost to AI will be offset by new ones, technologists warn of a “universal worker” capable of performing a wide range of tasks, potentially displacing significant portions of the workforce. The Financial Times reported that AI evangelists often view economists as overly conservative, relying on past trends rather than anticipating future breakthroughs.
The differing perspectives likewise reflect a broader pattern of technologists being bullish on AI’s economic impacts, while economists remain more reserved. As reported by TIME magazine, economist Anton Korinek is among those sounding the alarm about the potential societal transformations brought about by AI, a view not universally shared within the economics community.
SIEPR has not released a public statement summarizing the conclusions reached at its December 2025 forum. A follow-up meeting is scheduled for the fall of 2026 to reassess the economic impacts of AI as the technology continues to develop.
