AI Could Push 11 Million U.S. Workers Into New Careers by 2035, Study Finds | SOFX

TEXT START: About 11 million U.S. workers could need to switch occupations by 2035 as AI and automation reshape the labor market, according to a McKinsey Global Institute survey.

The Dissection

The article converts a displacement event into a mobility story. It foregrounds the reassuring aggregate: roughly 40 million jobs may emerge against 36 million in reduced labor demand. But the decisive facts are more damaging: 11 million workers may require entirely different occupations, 70% may experience role reinvention, and occupational churn could run at three to four times the historical rate.

“More jobs than workers” is treated as proof of resilience. It is not. Job counts do not establish that displaced workers can access the new jobs, earn comparable wages, retain bargaining power, or participate in production rather than merely remain attached to a labor market controlled by AI capital.

The Core Fallacy

The central error is treating labor demand as fungible. A newly created AI-sector job does not automatically absorb an automated worker. The relevant variables are ownership, skill thresholds, geography, credentials, wage levels, timing, and control of the productive systems.

The article also mistakes demographic scarcity for economic security. An aging workforce may reduce the number of available workers, but that does not prevent AI from making specific categories of human labor unnecessary. It merely disguises the transition in aggregate employment statistics.

Under the Discontinuity Thesis, the 40-million-job projection does not refute collapse. It may describe a transitional labor market in which AI creates output and specialized roles while severing the mass employment-to-consumption circuit for everyone else.

Hidden Assumptions

• New jobs will be accessible to the workers displaced by automation.
• Occupational retraining can occur at the projected pace without severe wage loss or prolonged unemployment.
• Jobs created by AI expansion will be broadly distributed rather than concentrated among owners and controllers of AI capital.
• “Role reinvention” preserves meaningful productive participation instead of reducing workers to supervised, precarious, or low-value residual tasks.
• Labor shortages caused by demographic aging will offset automation across the occupations that actually lose demand.
• The economy can absorb 770,000 occupational transitions per year without institutional failure.
• Gross job creation is an adequate proxy for income security, bargaining power, and social stability.
• Forecasts of 2035 demand are reliable enough to justify treating the transition as manageable.

Social Function

This is partial truth serving as transition management and ideological anesthetic.

The article accurately reports the approaching velocity of disruption. Its own figures admit that automation could erase demand for 36 million jobs and force 11 million people into new occupations. But the “40 million jobs” counterweight turns a distributional crisis into an employment spreadsheet. The phrase “the next decade’s challenge is mobility, not scarcity” redirects attention from who owns the machines to whether workers can move fast enough to remain useful to them.

The result is elite self-exoneration by arithmetic: if total jobs remain positive, no one must answer for the destruction of labor leverage, wage compression, or concentrated ownership.

The Verdict

The article does not demonstrate that the post-WWII economic order survives. It documents its lag phase.

The headline is a reskilling narrative wrapped around a structural warning. Aggregate employment may remain positive while productive participation collapses for large groups. The decisive question is not whether 40 million jobs appear; it is whether ordinary workers retain ownership, bargaining power, or indispensable control over production. On the supplied evidence, the article offers no answer—and its optimistic framing quietly assumes the answer is yes.

Under the Discontinuity Thesis: P1 is advancing, P2 is ignored, and P3 is being hidden behind gross job creation. The labor market is not being preserved. It is being refitted around the people who control the replacement machinery.