Microsoft posts Acemoglu's bearish AI forecast: 1.5% GDP growth over ten years

Acemoglu's forecast stands out because it comes from a Nobel Prize-winning economist and was published by Microsoft itself, a company with a major stake in AI adoption. His view contrasts sharply with some AI labs that predict far larger economic effects. The bottleneck, he argues, is not technology but human nature: companies must reassign tasks, upskill workers, and restructure before productivity gains kick in. This process could take longer than electrification did.

He also argues that bigger models won't solve this. What's missing, he says, are easy-to-deploy apps that change how things get made. And AI that extends human skills will beat full automation on productivity, since even 99 percent accuracy often isn't enough for last-mile problems and real user needs.

For Microsoft, this thesis is convenient: the company can bolt AI onto existing products without betting on large-scale automation. The article appeared in Microsoft's eccentric corporate blog "The Humanist Review of AI," where authors sign their pieces by hand. Whether Acemoglu's cautious view proves right may hinge on how quickly companies can adapt their workflows — a pace that remains uncertain.