AI Will Reshape 11 Million U.S. Jobs by 2035: McKinsey Report

Artificial intelligence (AI) and automation will trigger a historic shift in the American labor market over the next decade, potentially forcing roughly 11 million U.S. workers into entirely new occupations by 2035, according to a study released Tuesday by the McKinsey Global Institute.

While social media narratives often warn of widespread job losses, McKinsey’s findings suggest the nation’s primary economic challenge over the coming decade will be worker mobility rather than job scarcity. Driven by AI innovation, economic growth, and a shrinking labor pool caused by an aging population, the U.S. is projected to feature more overall job openings in 2035 than it does today.

Automation will eliminate or displace the equivalent of 36 million positions by 2035, according to McKinsey’s projections. However, expansion in technology, AI-driven sectors, and the broader economy is expected to offset those losses by generating demand for 40 million new jobs, a net gain of 4 million positions.

Of the 36 million workers affected by automation, an estimated 25 million will remain in their current occupations as industry growth absorbs the technological disruption. The remaining 11 million — representing about 6.5% to 7% of the current U.S. workforce — will need to change career paths entirely.

Depending on the speed of tech adoption, McKinsey estimates the total number of displaced workers needing occupational shifts could range from 6 million to 16 million.

“The next decade’s challenge is mobility, not scarcity,” researchers wrote, noting that the transition “may require the largest and most sustained workforce transformation in U.S. history.”

Meeting these demands will require approximately 770,000 American workers to switch occupational fields annually—nearly 3.6 times the historical average rate.

The report highlights a sharp imbalance in how technological displacement will affect the labor force. More than 75% of the workers forced to transition are concentrated in three key sectors – office and administrative support, retail and customer sales, and transportation and logistics.

Demographics also play a critical role. Lower-wage workers are 7.6 times more likely to face a complete career change than their higher-earning counterparts. Similarly, workers without a college degree are 1.8 times as likely to face occupational displacement compared to those holding a bachelor’s degree.

Navigating this transition will prove difficult for most displaced employees. McKinsey estimates that only 14% of affected workers have access to a “direct pathway,” a shift requiring minimal retraining without a drop in pay. Roughly 40% face a “winding pathway” requiring significant reskilling, while nearly half face an “unpaved pathway” blocked by steep skill gaps, strict educational requirements, or high credentialing barriers.

Credentials will be crucial, as McKinsey found that 85% of growing job roles require formal certifications or degrees. Meanwhile, employer demand for specialized skills is shifting rapidly: job postings requiring AI fluency skyrocketed 11-fold between 2022 and 2026, while demand for adaptability increased fivefold.

The report arrives as the U.S. labor market operates in a “low-hire, low-fire” holding pattern, characterized by low turnover and limited mobility. Recent Bureau of Labor Statistics data shows job openings dropping to a five-month low in August, while voluntary quits—a key barometer of worker confidence—remained near a six-year low.

The sluggish environment is weighing heavily on public sentiment. The Conference Board’s Consumer Confidence Index dropped 6.7 points in September to 81.9, hitting a 12-year low, while workplace community platform Glassdoor reported its Employee Confidence Index fell to a record low over the same period.

The issue is not whether AI leaves the economy with enough jobs on paper but whether workers can realistically move from the occupations losing demand into the ones gaining it, according to Stephanie Walter, practice leader for AI Stack & Enterprise Application Development at HyperFRAME Research.

“Those jobs may require different skills, credentials, wages, or locations, so 11 million occupational transitions would still be an enormous economic and organizational challenge,” Walter said. “For employers, workforce planning cannot be an HR exercise that happens after automation. It must be part of the AI deployment itself: identify which tasks will change, redesign roles, train people before displacement occurs, and measure redeployment as seriously as productivity. AI may not create mass unemployment, but a poorly managed transition could still create substantial disruption and inequality.”