Dario Amodei spent 2025 warning of massive white-collar job losses from AI. Now the Anthropic CEO is pivoting to a more optimistic narrative about job creation. The about-face reveals the uncertainty even AI leaders have about their technology's actual impact.
Last year, Amodei was clear-eyed about the disruption ahead. AI would automate knowledge work at scale. Lawyers, accountants, analysts—entire professions faced displacement. It wasn't a question of if, but when and how many. The warnings came from someone building the technology, making them hard to dismiss as Luddite fear-mongering.
Now? Amodei is invoking Jevons paradox—the economic principle that increased efficiency often leads to increased total consumption. Make something cheaper and faster, and you don't necessarily reduce demand for it. You expand what's possible, creating new jobs in the process.
According to Fortune, Amodei now argues that AI will make white-collar work so efficient that it unlocks entirely new categories of economic activity. The jobs won't disappear—they'll transform. Instead of fewer lawyers, we'll have more legal services accessible to more people. Instead of fewer analysts, we'll have deeper analysis across more domains.
It's a convenient narrative shift, particularly as AI regulation looms and companies face increasing scrutiny over their societal impact. I've watched too many founders change their tune when the political winds shift. The question is whether this represents genuine evolution in thinking or strategic messaging.
To be fair, Jevons paradox is real. When calculators became cheap and ubiquitous, we didn't eliminate accounting jobs—we expanded what accountants could do. But calculators didn't replace accountants. They augmented them. The AI we're building now is designed to replace entire categories of cognitive labor.
The honest answer is that nobody—not even the people building these systems—knows exactly how this plays out. Amodei's initial warnings were educated guesses based on capability trajectories. His current optimism is equally speculative, just pointing in a different direction.
What changed? Maybe new data about how companies are actually deploying AI. Maybe conversations with economists who study technological transitions. Or maybe just the realization that doom predictions aren't great for business when you're trying to sell AI services to those same white-collar workers.
