Old Playbooks, New Disruption: What History Can (And Can’t) Teach Us About AI and Work

In May, OpenAI’s Sam Altman launched the company’s Industrial Policy for the Intelligence Age, calling for a new social contract on the scale of the New Deal. Weeks later, speaking to a Sydney banking conference, he acknowledged the entry-level white-collar losses he’d braced for hadn’t materialized, and said he no longer expected a “jobs apocalypse.”

During his confirmation hearings in April, now-Fed Chair Kevin Warsh had called AI “the most disruptive moment in modern economic history,” even while conceding he was more confident that AI would raise productivity than about when those gains would reach workers. 

Last summer, Amazon CEO Andy Jassy, in the memo that preceded the company’s largest-ever round of corporate layoffs, put it plainly: “It's hard to know exactly where this nets out over time.” 

Many players in the economy—across sectors and industry—expect AI to reshape work in ways that lack modern precedent. But even within that growing group, there’s no agreement on how far the shift will go, how fast, or what to do about it, if anything. However, government will have to decide what to do—or not do—before the picture is clear, making the risks highly asymmetric. And while there are gaps of knowledge, the window for decision-making might be closing—and the Future of Work cannot wait on certainty.