AI Just Eliminated an Entire Department at a Fortune 500 Company: The CEO Called It ‘Necessary’

No big announcement was made. No all-hands meeting at a podium with a somber CEO. For a whole department, nothing more than an email and a calendar hold. Within weeks, the floor where about two dozen analysts had been employed was discreetly moved. Additionally, the CEO of the company gave a succinct response when questioned about it during an investor briefing: it was essential.

It reads like a headline from five years in the future. However, in the spring of 2026, this is taking place within some of the world’s most well-known corporations. AI didn’t just appear in the business world; it was publicly announced, hotly debated, and somehow arrived before most people were prepared. Almost overnight, the difference between what executives said in conference rooms and what is currently happening on organizational charts has vanished.

AI will replace “literally half of all white-collar workers” in the US, according to Ford Motor Company CEO Jim Farley, who made this clear at the Aspen Ideas Festival last July. Some dismissed it at the time as CEO hyperbole, which is a provocative statement meant to create buzz. However, those who have been observing the real workforce data, internal restructuring memos, and the abrupt halt to some hiring categories at large companies were not shrugging. They were keeping track.

You wouldn’t notice much of a difference if you strolled by the type of corporate campus where these changes take place. The same hours see the parking lots fill up. The scanners for badges beep. The coffee is still awful. However, a group of AI agents operating silently on servers you’ll never see has taken the place of a team that used to spend its days processing reports, directing customer escalations, or creating internal analyses somewhere on the fourth floor. One LinkedIn executive recently described the math as “brutal.”

The specificity of this moment sets it apart from all previous waves of automation anxiety. Unlike in the 1980s, workers on manufacturing floors are not being replaced by mechanical arms. Call centers are not relocating offshore as they did in the 2000s. This time, the work is knowledge-based, white-collar, credentialed, and comes with a business card, a health plan, and a sense of permanence. Departments that previously relied on cognitive labor to justify their existence—synthesizing, making decisions, and communicating—are being reorganized around systems that can accomplish all of that, at scale, without the need for performance reviews or lunch breaks.

Perhaps the best example of what this looks like when it works is JPMorgan Chase. Over 200,000 workers have access to the bank’s GenAI toolkit, which is internally referred to as the LLM Suite. According to reports, one of its apps, Sales Assist, increased gross sales per user by about 188% between 2022 and 2023. That program isn’t a pilot. The way the company’s employees generate value has fundamentally changed as a result. It’s also important to note that JPMorgan isn’t firing employees just yet. It’s increasing the potential of every individual. However, anyone who is willing to do so can see the implication—the awkward math that lies just below the productivity figures.

Researchers at MIT Sloan and Harvard Business School have been closely examining this shift, and their conclusions don’t exactly encourage relaxation. For nervous employees, the high failure rate of enterprise AI projects—roughly 80%—sounds like good news. Really, it isn’t. In reality, it represents a leadership issue rather than a technological one. Today, value is being lost by executives who are slow to adjust, slow to guide, and slow to redesign workflows around AI’s true capabilities. The pressure from the competition ensures that they will catch up, and when they do, the restructuring will be quicker and more aggressive rather than more gentle.

Some CEOs seem to be saying in public what they have been saying in private for at least two years. The CEO of Anthropic cautioned in May of last year that within five years, artificial intelligence might eliminate about half of all entry-level white-collar jobs. For the head of an AI company to say that is remarkable; it sounds more like a confession than a prediction. And yet here we are, witnessing it start to appear in headcount reports, quarterly filings, and the more subdued announcements that don’t make the front page.

The analyst in her mid-thirties who recently discovered that her entire function is being absorbed into a platform she has never touched is one of the people caught in the middle of this, not the executives portraying it as progress or the AI researchers celebrating capability benchmarks. The professional whose career path was based on proficiency in precisely the type of work that is most automatable. Businesses like Moderna have made a concerted effort to do this by implementing AI to lessen laborious legal review tasks, educating staff members on new tools, and redefining the technology as a career accelerator. That is conceivable. It’s not universal either. Not every company has the patience or the culture to support it.

Depending on your stance on the restructuring, this could be remembered as either a productivity revolution or a labor catastrophe. In the limited financial logic of competitive markets and shareholder expectations, the CEO who deems it necessary is most likely correct. That does not, however, make it simple. Furthermore, it doesn’t address the more general question of what happens to the people left behind by the math, which is still very much open and worthwhile.