The megamanager era: AI is doubling bosses’ workloads—and the costs are just beginning to show

Call it the megamanager era. Driven by AI-enabled cost-cutting, leaner bureaucracies, and a relentless corporate push to rationalize headcount, companies have spent the past three years gutting their middle-management ranks, leaving whoever survives with a dramatically larger portfolio of people. The data is as official as it gets, coming straight from the Bureau of Labor Statistics. The average number of a manager’s direct reports has nearly doubled since Gallup began tracking the figure in 2013.

If AI can handle scheduling, summarize performance reviews, monitor project timelines, and surface early warning signals about team dysfunction, do you really need as many human coordinators? Meta’s new applied AI engineering division has taken the logic to its most aggressive extreme, deploying a 50-to-1 employee-to-manager ratio—roughly double what was once considered the outer limit of a functional organizational structure. Whether the rest of corporate America follows that example or it becomes a cautionary tale may define the future of work for the next decade.

The pros: speed, savings, and structural clarity

For companies, the immediate math looks appealing. Fewer managers mean lower headcount costs, flatter hierarchies, and (in theory) faster decision-making. When a senior vice president no longer has to relay information through two or three layers of middle management before it reaches the people doing the actual work, information can travel faster, and accountability can land closer to the front lines. A 2024 Gartner analysis predicted that one in five businesses plan to use AI specifically to streamline organizational layers.

AI is also genuinely helping some managers cope with the expanded workload. Tools that automate administrative tasks—flagging performance issues, synthesizing team data, drafting communications, and coordinating schedules across large groups—are reducing the friction that once consumed hours of a manager’s week. Done well, this kind of AI augmentation could make the megamanager model viable: a skilled, well-supported boss leading a dozen people might be more effective than a distracted, paper-buried boss leading six.

The productivity case has deep historical precedent. A sweeping analysis published this week by Morgan Stanley looked at five prior American innovation waves — from the first Industrial Revolution through the internet—and found a consistent pattern: transformative technologies raise output per worker, particularly when paired with deliberate organizational redesign. Chief U.S. economist Michael Gapen’s team found that electrification doubled output per hour in nonfarm business between 1900 and 1929. The internet accelerated labor productivity growth from roughly 1.5% per year to nearly 3.0% per year by 2000. AI should follow the same arc, Gapen suggested—with one critical caveat. Those productivity gains have historically materialized years, sometimes decades, after the initial disruption, not simultaneously along with it. The pain tends to come first.

The human ledger is looking considerably worse than the balance sheet. Another Gartner survey found 75% of HR leaders believe managers are already overwhelmed by their expanding responsibilities, and 69% say managers lack the skills to lead change effectively even before full AI integration takes hold. Gallup data show that global employee engagement has fallen to just 21%, near a 15-year low, with managers themselves—not just the people they supervise—reporting some of the sharpest drops in workplace satisfaction of any cohort. The Wall Street Journal recently argued that work is increasingly “joyless” as many offices take on a funereal atmosphere in the age of the megamanager.

Perhaps the most underappreciated cost of span-of-control inflation is what happens to the people at the earliest stages of their careers. Coaching, mentorship, and hands-on development — the soft infrastructure that has historically built management pipelines and transmitted institutional knowledge from one generation to the next—are the first casualties when a single boss is stretched across 12 people rather than 6. A manager with a dozen direct reports simply cannot spend the same number of hours per person nurturing potential, giving real-time feedback, or advocating for junior employees in rooms they’re not in. That gap accumulates, posing a threat to talent development.

Neil Thompson, a research scientist at MIT who studies how AI capabilities evolve across the economy, offers a more nuanced frame for understanding what’s actually at stake. In his research—which evaluated 40 AI models across thousands of real-world job tasks, each assessed by practitioners in the relevant field—Thompson and his colleagues find that automation doesn’t affect all parts of a job equally. The critical variable is whether the tasks being automated are the expert parts of a role or the administrative scaffolding around them.

“If part of your job gets automated and it’s something that really didn’t use the expertise that you needed, that’s great,” Thompson said. “You get to spend more of your time on the part of your job that is really valuable.” His research, co-authored with MIT economist David Autor, finds that when automation eliminates the lower-expertise components of a job, wages for the remaining workers actually tend to rise: there are fewer of them, but they’re doing more of what makes them irreplaceable. The danger, Thompson warns, is the opposite scenario: when AI targets the expert core of a role — the way GPS wiped out the navigational mastery that once defined a taxi driver’s craft—wages fall, and the profession’s identity hollows out.

That warning echoes across a century and a half of economic history. Every major innovation wave in American history—from steam power and railroads to electrification to the internet—displaced workers, concentrated early gains among capital holders, and provoked political backlash before productivity benefits eventually broadened. Morgan Stanley’s economists note that “workers were reallocated rather than rendered obsolete” across all five prior waves—but the transition periods were wrenching, and the distribution of benefits depended heavily on policy choices, investment in education, and institutional adaptation. When those systems responded well—as they did during the mid-20th century’s “Great Compression,” which coincided with expanding unions, progressive taxation, and the GI Bill—innovation produced broadly shared prosperity. When they lagged, inequality deepened.

“Since 1980, income and wealth concentration have risen sharply, driven by returns to capital, skill-biased technical change, and public policy choices that reversed Great Compression-era policy,” Gapen’s team wrote. “Innovation itself does not predetermine inequality: institutions and public policy mediate how gains are distributed.”

Goldman Sachs economists estimate AI has so far raised the overall unemployment rate by just 0.1 percentage point—a modest headline figure that obscures a bifurcated picture: jobs easily substituted by AI are contracting, while roles augmented by AI are actually growing. The radiologist’s case is the most instructive example on offer. When Geoffrey Hinton, the godfather of deep learning, predicted in 2016 that AI would replace radiologists within five years, it seemed like an obvious forecast. Instead, as Axios noted on the complex adoption picture, radiologists have broadly adopted AI tools, used them to read more scans more accurately, and have seen both their numbers and their pay increase since. The technology didn’t eliminate the profession. It redefined it.