We’re only about five months away from an important milestone here in the emerging age of artificial intelligence. Back in February of this year, Microsoft AI boss Mustafa Suleyman, also a cofounder of Google DeepMind (formerly DeepMind before its acquisition by the tech giant), made this prediction in an interview with the Financial Times: “White-collar work, where you’re sitting down at a computer, either being a lawyer or an accountant or a project manager or a marketing person—most of those tasks will be fully automated by an AI within the next 12 to 18 months.”
Now, an economist or even a well-informed FT reader—knowing full well that a job is a bundle of tasks and that AI automation of specific tasks doesn’t automatically mean a worker loses their job—might not have read that forecast as one predicting massive white-collar unemployment in a year to a year and a half—now just five to eleven months away! But for the less sophisticated, that’s exactly what Suleyman was predicting, much to his exasperation, as indicated in this Verge interview conducted by the publication’s editor in chief, Nilay Patel:
[Patel]: This is what [you] said to the Financial Times four months ago: “White-collar work when you’re sitting down at a computer, either being a lawyer or an accountant or a project manager, or a marketing person, most of those tasks will be fully automated by an AI within the next 12 to 18 months.”
That’s four months ago. That implies that a year from now, lawyers, accountants, project managers, and marketing people will not have jobs. Their jobs will be automated. Is that still your timeline?
[Suleyman]: No, no, no. Hold on a sec. So I said “tasks” in the quote that you’ve just said. I said tasks. So that does not mean jobs. It’s a very important distinction. In labor economics, there is an entire taxonomy of sub-components of a role or a function in an organization. . . .
That does not necessarily mean that the role goes away at all.
Alas, America is populated by the economically uninformed and social media headline readers. So Suleyman’s prediction inevitably fed into a public perception of an impending “white-collar bloodbath,” as alarmingly described in an Axios headline from May 2025. And no doubt that perception currently feeds into public hostility toward AI and the data centers that enable it. As such, I can understand the temptation to assume that dealing with AI disruption will be the dominant political issue over the next two years leading up to the 2028 presidential election.
But I’m skeptical. First, even the lesser Suleyman forecast would be a considerable jump from today, when AI can do some small fraction of unsupervised, white-collars tasks across the economy. But imagine a world two years from today, where even as ever-improving AI automates lots of tasks, it also allows workers to spend more time on others, or even creates new things to do and entirely new jobs—and that’s not even counting jobs created indirectly, such as by the AI infrastructure build-out. In this entirely plausible world, perhaps electoral politics will focus on the economic condition suggested in this chart (and today’s CPI report):