AI Won't Destroy Jobs, But Millions Will Need to Learn a New Trade 2026

AI won't destroy the job market outright, but millions of employees will need to learn a new trade, according to a new McKinsey report. The findings echo the spreadsheet revolution of the early 1980s, when accountants feared for their jobs but ultimately adapted.

The report estimates that by 2035, AI-driven automation will reduce labor demand by the equivalent of 36 million jobs in the US alone. However, this figure represents hours of work no longer needed due to time savings, not actual layoffs.

AI Automation: A Repeat of the Spreadsheet Revolution?

In the early 1980s, accountants saw their jobs threatened by a new tool: the spreadsheet. Tasks that once took a day were completed in seconds. Manual calculation departments vanished, but accountants didn't; they moved into other roles.

Nearly half a century later, AI reopens that fear, but this time the threat targets the entire labor market at once. McKinsey's report attempts to measure that impact and arrives at conclusions similar to what accountants discovered at the dawn of the computer age.

McKinsey's Numbers: 36 Million Jobs Affected, But Not Lost

McKinsey calculates that automation will reduce demand equivalent to 36 million jobs by 2035. Of those, about 25 million are offset within the same profession. That means tasks change, but the person stays employed.

However, 11 million of those jobs—equivalent to 6.5% of the US active population—would require retraining and switching to new occupations. This is the real challenge: not mass unemployment, but massive reskilling.

• 36 million jobs in equivalent hours will be automated by 2035.
• 25 million of those are absorbed within the same profession through task shifts.
• 11 million workers (6.5% of the active population) will need to learn a new trade.
• The spreadsheet revolution of the 1980s offers a hopeful precedent for adaptation.

Why This Time Is Different—and Why It Isn't

Unlike the spreadsheet revolution, AI affects a broader range of industries simultaneously. From customer service to legal analysis, no sector is immune. But the mechanism is similar: automation handles routine tasks, freeing humans for more complex work.

The key difference is speed. The spreadsheet took decades to permeate; AI is advancing in months. That leaves less time for workers to adapt organically, making proactive retraining essential.

What the Numbers Really Mean for Workers

The 11 million workers who must switch occupations face a daunting task. They will need new skills, certifications, and often a change in mindset. Yet the report suggests this is manageable if addressed early.

For the 25 million whose jobs are transformed but not eliminated, the challenge is continuous learning. Tasks will evolve, and workers must evolve with them. Companies that invest in upskilling will retain talent; those that don't risk disruption.

The Reskilling Imperative: Who Pays?

Who bears the cost of retraining 11 million workers? The report doesn't specify, but history suggests a mix of employer investment, government programs, and individual initiative. The spreadsheet era saw companies retrain staff internally; today, the scale demands broader collaboration.

Policymakers must create incentives for reskilling, while educational institutions need to offer flexible, fast-track programs. Workers themselves must embrace lifelong learning as a career norm, not an exception.

What to Watch Next

As AI adoption accelerates, watch for signals of labor market stress: wage stagnation in automatable roles, rising demand for technical skills, and gaps in retraining access. The transition will be uneven across regions and industries. The background to this story is set out in Business: latest articles.

The lesson from accountants is clear: technology doesn't erase professions, but it reshapes them. The winners will be those who adapt early, and the losers those who wait. For millions, the future depends on learning a new trade—not because AI destroys jobs, but because it changes them. A related thread runs through Matt Damon's The Last Duel: The 2021 Armor Movie You Missed. Earlier coverage in AI Won't Destroy Jobs, But Millions Will Need to Learn a New Trade shows how the argument developed.