Daron Acemoglu, an MIT economist who won the Nobel Memorial Prize in Economic Sciences in 2024, estimates that artificial intelligence will deliver roughly 0.55% in total factor productivity gains over the next decade. He also estimates that about 5% of tasks will be profitably automated in the near term, a figure he stated is equivalent to a 1% or 1.5% increase in GDP.

Acemoglu stated that about 20% of current AI discourse is intellectually serious, clarifying that he considers much of the rest speculative or close to fictional rather than stupid. "I find all of this discussion of capitalism so brainless. That's what we should be talking about. What we should be talking about is the displacement and unequalizing roles of AI," Acemoglu said. He noted that "a lot of the left is a big contributor" to the speculative discourse.

Acemoglu criticized the phrase "colonizing AI" as unhelpful Marxist rhetoric and stated that he does not like the term "capitalism." "I don't like the term capitalism. It makes it sound like there is a uniform model that includes Sweden, Egypt, Argentina, Honduras, the United States, South Korea, Japan. There's no overlap between these economies, how they are organized," he said. He stated that the only overlap he sees between various national economies is that they have markets, noting that the Soviet Union also had markets.

Acemoglu, who developed the framework of inclusive versus extractive institutions with co-author James Robinson in books including *Why Nations Fail* and *The Narrow Corridor*, argued that today's AI hyperscalers fit the extractive institutional mold. He characterized them as having concentrated ownership, regulatory capture, and a business model that extracts data and attention at scale.

Regarding productivity, Acemoglu stated that gains from automation only materialize if machines can do tasks significantly cheaper or better than humans. He said, "It's not that you cannot get big productivity gains from automation. It is that it's not as easy as sometimes it's presumed." He added that true "human complementarity" would be AI that enables workers to do things they simply couldn't do before, and noted that most research on AI productivity is overblown because it focuses on easy, well-defined tasks where context is clear.

Acemoglu stated that generally huge productivity gains from automation require something close to artificial general intelligence. "So that's why AGI is not just a theoretical issue — it's really relevant for these productivity projections," he said. He argued that current AI models perform badly across too many dimensions of real-world work, stating they cannot read a room, connect non-obvious dots across domains, or succeed where human judgment is most valuable.