Note: See here for the article on How Limited Partners Can Review Syndicator Reports and Assert Their Rights.
There is a version of this story everyone has already read.
Tech layoffs are rising. AI is the culprit. The robots are coming for white-collar work, and the numbers prove it. Though the numbers are undeniable, I think their implication is far more nuanced.
By early May 2026, more than 100,000 tech workers had reportedly been laid off. Q1 layoffs were up sharply from the same period last year, the steepest increase since the post-pandemic cuts of 2023. Amazon and Oracle both announced large reductions. By early June, some layoff trackers showed the pace approaching nearly 1,000 tech jobs lost per day.
That sounds like an AI story. It may be, eventually. But for now, it’s more like a cost story with AI used as a scapegoat. Companies have learned to frame layoffs in the language of technology investment. They cite efficiency, restructuring, and the need to build for an AI-driven future. The cuts may be driven by margin pressure, overhiring, or slower growth, but AI gives the decision a more forward-looking explanation.
It is a convenient narrative. It is also incomplete.
We see this up close through our work with clients every day. AI is absolutely changing how work gets done. Developers are using it to move faster. Teams are using it to test ideas, generate drafts, clean up data, summarize information, and reduce the amount of manual work sitting between a question and a usable answer.
That is not nothing. In many cases, it changes the economics of a project. But it is not the same as replacing the people responsible for the work.
The AI tools are useful, sometimes surprisingly useful. They can accelerate the first pass. They can turn messy inputs into something structured. They can help a developer troubleshoot faster or help a team explore options before committing resources. What they do not do, at least not reliably, is own the business judgment, the client context, the final decision, or the accountability for getting the thing right. That’s a distinction that has gotten lost in the layoff narrative.
In most companies, AI has not yet taken over work in a way that allows entire functions to disappear cleanly. The more common pattern is less dramatic. Teams are being asked to do more with fewer people, using AI as leverage. A task that once took five hours may now take two. A first draft may take minutes instead of a day. A data pull that once required a junior analyst may now start with a prompt and end with human review.
That changes productivity. It changes staffing assumptions. And what entry-level work looks like. But it does not fully explain the current layoff wave.
Tech companies hired aggressively during the pandemic. Demand was high, capital was cheap, and growth was treated as a mandate. Companies added engineers, product managers, recruiters, operations teams, and layers of management to support a world that did not last.
Then interest rates started rising. Growth slowed. Investors stopped rewarding headcount expansion and started asking about margins. The same payrolls that looked reasonable in 2021 became hard to justify in 2026.
AI did not create that problem, but it’s giving companies an easy out to resolve it. Saying “we overhired when money was cheap” is an ugly admission. Saying “we are reallocating resources toward AI” sounds strategic. It tells shareholders the company isn’t retreating, but evolving.
That is the center of the current moment.
Many companies are not cutting jobs because AI already replaced those workers. They are cutting jobs to free up capital for AI infrastructure, AI talent, and AI-related investment. The layoffs are often funding the AI buildout rather than proving the AI buildout already worked.
We see the same distinction in practical technology projects. AI can make a team faster, but speed alone does not redesign an organization. A good internal tool can reduce repetitive work. A better workflow can remove friction. An AI-assisted process can save hours. But those gains still need humans to define the problem, check the output, manage exceptions, and understand what the business is actually trying to accomplish.
That is where the public conversation gets sloppy.
There is also a second pattern: companies making labor decisions based on what they believe AI will soon be able to do. Management looks at a function, assumes the tools will improve quickly, and cuts ahead of the curve. The company may not have an AI system that can fully absorb the work today. It may simply be betting that the gap will close.
Sometimes that bet will pay off. Sometimes it will not. When it does not, the correction is usually quiet. Roles get backfilled later, often offshore or at lower cost. Contractors appear. Internal teams absorb work they were never staffed to handle. The public explanation does not get updated. AI took the blame, and the savings remain. None of this means AI is irrelevant to employment; in some industries, especially data-heavy ones, AI is already doing more than assisting. Financial services, logistics, insurance, and parts of healthcare have functions where large volumes of information need to be processed, classified, flagged, and routed. In those settings, AI can reduce the number of people needed to perform work that was previously manual, repetitive, and expensive. That is closer to actual displacement.
But even there, the change is uneven. AI performs well in bounded systems with clean inputs, defined outputs, and strong oversight. It performs less reliably when the work requires judgment, accountability, relationship management, or context that is difficult to encode.
Inside most companies, AI is still closer to a very fast junior employee than an independent operating layer. It can help produce the work, but can rarely own the result. That is why the layoff numbers need to be read carefully.
A company can cut 5,000 jobs and announce a major AI investment in the same quarter. That does not mean the AI replaced 5,000 people. It may mean the company is shrinking payroll, redirecting capital, satisfying investors, and preparing for a future it does not yet know how to run. The actual employment risk will sharpen when AI moves from tool to infrastructure.
A tool helps employees complete tasks. Infrastructure changes the design of the company, determining how work moves, who reviews it, where decisions are made, and how many people are needed between input and outcome. That shift has not happened broadly yet.
When it does, the signs will look different. Companies will do more than announce efficiency programs. They will redesign departments around AI systems. Workflows will be rebuilt instead of merely accelerated. Entry-level roles will narrow or disappear in certain functions. Middle-management review layers may shrink. Accountability will become harder to locate because more decisions will pass through systems rather than people. That is the employment story worth watching.
Today’s tech layoffs are being described as AI disruption. In some cases, that description may fit. In most cases, it is doing more work than the technology itself. From what we see in actual technology work, AI is powerful, useful, and increasingly hard to ignore. But it is still mostly leverage. It helps good teams move faster. It does not automatically make the team unnecessary.
For now, AI is not the whole explanation. It is a productivity tool, a capital priority, a market signal, and a useful excuse. When it becomes the main driver of workforce replacement, we will not need a layoff memo to tell us.
This material has been prepared for informational purposes only, and is not intended to provide or be relied upon for legal or tax advice. If you have any specific legal or tax questions regarding this content or related issues, please consult with your professional legal or tax advisor.
