The Macro Data Is Right. The People Still Aren’t Ready.

Greg Ip’s argument in the Wall Street Journal is solid, and we should take it seriously. Technology has never caused a sustained job apocalypse in the United States. Software developer employment is up 5% year-over-year. Computer science graduates are earning a larger wage premium than they did fifteen years ago. Business software spending jumped 11% in Q4 2025. The macro indicators don’t show a workforce in collapse. They show a workforce in transition.

And that distinction matters enormously for anyone leading people through it.

Ip is answering a macroeconomic question—will AI cause a recession through mass unemployment? His answer, backed by Labor Department data and historical precedent, is almost certainly no. We agree. The economy has absorbed technological disruptions before, from textile automation to ATMs to spreadsheet software replacing bookkeepers while creating more accountants and financial analysts than it displaced. The overall system adapts. It always has.

But aggregate employment data doesn’t walk into your office on a Tuesday afternoon, convinced their career is over. Macro trends don’t sit across the table from you during a performance review, struggling to articulate skills they’ve spent twenty years building that suddenly feel irrelevant. The economy adjusts. Individual humans—working under stress, carrying mortgages, managing families—adjust much more slowly and painfully than any chart suggests.

Thanks for reading Lead from the Front! This post is public; feel free to share the knowledge.

This is the gap we keep seeing in the AI conversation. The economists are talking about labor markets. The technologists are talking about capability curves. Almost nobody is talking about the actual mechanism by which a 45-year-old marketing director becomes proficient with tools that didn’t exist eighteen months ago. And that mechanism is where leadership either steps up or falls short.

Ip cites the bookkeeper-to-accountant transition as evidence that displaced workers find new, often better roles. We’d push on the timeline. That transition took roughly a decade. The bookkeepers who successfully became financial analysts didn’t do it because the market economy functioned correctly in the abstract. They did it because someone—a manager, a mentor, an institution—helped them bridge the gap between skills owned and skills needed. The invisible hand doesn’t retrain anyone. People do.

What Ip’s analysis misses isn’t the economics. It misses the implementation. And implementation is where organizations consistently fail, whether we’re talking about AI adoption, cultural change, or leadership development. Our research on Implementation Science shows that only 20% of evidence-based interventions achieve sustained real-world adoption. The knowing-doing gap isn’t a footnote to the AI transition. It is the AI transition, at least from inside any organization trying to navigate it.

We work with mid-market companies where this is playing out right now. The C-suite reads articles like Ip’s and feels reassured—the macro picture is fine, no apocalypse coming. Meanwhile, their middle managers are fielding daily questions from teams who are genuinely uncertain whether their roles exist in two years. Both perspectives are correct. And the leaders who treat this as purely an economic question rather than a human development challenge are going to lose their best people to organizations that take the transition more seriously.

Amy Arnsten’s research on stress and prefrontal cortex function offers practical insights into this moment. When people feel threatened—and career uncertainty is a threat response, regardless of what the employment data says—their capacity for learning new skills degrades significantly. The prefrontal cortex, which handles complex reasoning and skill acquisition, gets overridden by the amygdala’s fight-or-flight response. You can’t upskill a workforce that’s operating from a stress response. You have to address the psychological reality first, then build capability systematically.

That means leaders need to do three things most organizations aren’t doing yet. First, run an honest skills audit—not a generic competency framework, but a clear-eyed assessment of what your people can do today versus what the work will require in eighteen months. Second, build transition pathways that look more like the extended apprenticeships that actually produce durable capability than like the three-day training programs that produce satisfaction surveys and little else. Phillippa Lally’s research at University College London found that genuine habit formation takes an average of 66 days, and complex professional skills take considerably longer. Third, measure whether people are actually adopting new capabilities under real working conditions, not whether they completed a course. Leading indicators—connection quality between managers and teams, psychological safety scores, and coaching frequency—will predict who navigates this transition successfully long before any lagging employment data confirms it.

Ip is right that the doomsday scenario requires a breakdown in the functioning of market economies. We’d add that the human scenario—the one playing out in conference rooms and one-on-ones across the country right now—requires a breakdown in how organizations develop their people. And that breakdown has been the norm, not the exception, for decades. Seventy percent of organizational change initiatives fail, according to McKinsey’s research. Only 11% of executives believe their leadership development programs achieve sustained results.

The macro data will take care of itself. The economy adapts. The real question for leaders isn’t whether AI will cause an apocalypse. It’s whether your organization can develop people fast enough, and well enough, to capture the opportunity Ip correctly identifies—before your competitors do it first.

Original Article: Ip, Greg. “Tech Has Never Caused a Job Apocalypse. Don’t Bet on It Now.” Wall Street Journal, February 27, 2026. https://www.wsj.com/economy/jobs/tech-has-never-caused-a-job-apocalypse-dont-bet-on-it-now-d192b579