Companies keep cutting jobs in the name of efficiency. A new survey suggests the work isn’t disappearing.

Korn Ferry’s Workforce 2026 report, released in September and based on a survey of more than 16,000 professionals across 11 global markets, found that 62% of respondents said their workloads had increased significantly over the past two years, and 61% said they were performing the responsibilities of more than one role. Nearly half, 45%, said they were too busy to deliver meaningful results that contribute to growth.

The findings land as companies continue to flatten their organizations. Last month, Uber said it would cut about 10% of its workforce, or roughly 3,300 employees, and reduce its number of managers by 20%. CEO Dara Khosrowshahi framed the restructuring as a way to create “clearer ownership, faster decisions, and more time spent building rather than coordinating,” and said it would reinvest the savings in growth and innovation. Unlike some tech executives, he did not attribute the cuts to AI.

Korn Ferry’s data raises a question about the other side of that efficiency push: Are companies eliminating jobs faster than they are eliminating the work? The consultancy calls the phenomenon the “two-job job”—employees who say they are covering the responsibilities of more than one role, sometimes while also managing AI on top of their regular tasks.

Managers who remain are overseeing more people than ever, and they’re feeling the squeeze. Korn Ferry found that 42% of organizations had cut management roles over the past year. More than half (55%) of the managers still in place said they were exhausted, and 39% of workers said a lack of managers left them feeling directionless.

At least one company appears to be rethinking how far to go. Fortune reported on Sept. 12 that Meta had begun asking some individual contributors in its Applied AI division whether they wanted to move back into management roles, according to Business Insider—a partial reversal after the company spent much of the past year stripping out management layers.

Eliminating jobs vs. eliminating employees

Peter Cappelli, the George W. Taylor Professor of Management at the Wharton School, questioned whether cutting managers delivers the productivity gains companies expect.

“It is not more efficient in terms of productivity to cut managers,” Cappelli told Fortune. Managers, he said, solve problems that help their teams get work done. “It just cuts costs and especially headcount, which is how CFOs in particular keep score.”

Cappelli also distinguished between eliminating jobs and eliminating employees. “The same work has to be covered by the remaining employees,” he said. “It is in fact a misnomer to say that jobs are cut. It is employees that are cut.”

For many workers, AI isn’t making the job easier. Korn Ferry found a wide gap between the C-suite and the front lines: 79% of CEOs reported improved efficiencies from AI, compared with 51% of individual contributors. Among workers the firm described as “AI-weary,” 52% said using the technology had increased their workloads.

Academic research points in a similar direction. A July working paper by Arvind Karunakaran of Stanford University, Kate Kellogg of MIT Sloan, and Batia Wiesenfeld of New York University’s Stern School of Business, based on a two-year field study of an academic medical center and a corporate law firm, found that building organization-wide generative AI tools created substantial work beyond employees’ day-to-day jobs. That work included trial-and-error experimentation, reviewing and refining outputs with colleagues across departments, and continually adapting solutions as AI models evolved.

Much of that work remained invisible to leadership, according to MIT Sloan’s Sept. 9 press release on the research. At the law firm, more than 80% of domain experts eventually dropped out of AI innovation efforts as the burden of experimentation mounted and the firm offered limited ongoing support. It ended up with three organization-wide AI tools in use. The medical center, which built AI work into job responsibilities, performance reviews, and promotions, has 141.

“Leaders often assume AI experimentation is a stretch assignment that motivated employees will absorb on top of their regular responsibilities,” Kellogg said in the release. That works only temporarily, she added: “Without meaningful support, recognition, and resources, people eventually disengage,” making it harder for companies to realize the value they expected from AI.

Korn Ferry argues that leaders need to rethink how work is divided between people and AI, and its motivation data shows the stakes. Employee motivation fell from 71% in 2024 to 61% in 2026.

“Growth doesn’t come from just asking people to do more,” Lesley Uren, CEO of Korn Ferry Consulting, said in releasing the report. She said cost-cutting and restructuring can produce savings without generating the kind of employee passion that drives productivity and growth.

For companies, the challenge is figuring out whether “lean” means removing unnecessary work—or simply asking fewer people to absorb more of it.

This story was originally featured on Fortune.com