AI Layoffs Are Buying the One Thing Every Competitor Will Have Too

10,970 of the 33,429 U.S. jobs cut in July 2026 were cut with AI named as the reason. That is 33% of the month's total, and it made July the 5th consecutive month in which AI layoffs led every other stated cause of American job cuts, according to Challenger, Gray & Christmas. No technology factor has ever held the top of that list that long.

Year to date through July, 112,713 cuts carry AI as the reason, 24% of every job eliminated in the U.S. this year. Since Challenger began tracking the category in 2023, the running total is 184,538. Those numbers deserve a slow reading, and the most careful reader so far turns out to be the firm that published them. Andy Challenger, the firm's Chief Revenue Officer, put it this way:

Naming AI in a layoff announcement can win over investors while pushing current and prospective employees away.

A labor-market executive is describing a communications decision with a price attached to it. Some share of these announcements is written for the people who own the stock, and the people who do the work read the same announcement.

AI is doing work in the press release that it may not be doing in the building

Some of those 10,970 cuts sit on top of real systems doing real work. Amazon shutting down Mechanical Turk was one visible instance of that kind of change, a task marketplace closing because the tasks moved. Others sit on top of a decision that was going to happen anyway, with a better word attached to it. From the outside, both land in the same monthly table, which is why the run of 5 months is worth reading slowly rather than as a verdict on what AI can do.

Everyone will have the same AI. Not everyone will have the same people.

The distance between the best AI system and the one a competitor can rent by the month keeps shrinking. When both sides of a market rent the same intelligence, the rented part cancels out of the competition. It lifts every competitor's floor at the same moment, which is a fine thing for the economy and a poor thing to call an edge. Subtract the identical AI from two companies and look at what each one still owns: accumulated judgment, tacit process knowledge, the people who can tell when the model's output is wrong. Cutting those people to fund the AI spends the one advantage that walks out the door when you cut it, in exchange for the one thing every competitor will shortly have too.

In March 2007, Circuit City fired 3,400 of its highest-paid, most experienced salespeople to bring costs down. The company was in Chapter 11 by November 2008. Best Buy hired a number of the people Circuit City had let go. The layoff did not save Circuit City, and the experience it released went to the competitor that survived.

The cuts are landing hardest on the people who evaluate the output

Through July, the year-to-date AI-attributed cuts fall unevenly across industries.

Technology alone accounts for 149,023 cuts this year, 31% of the full annual cut total across every industry. I track these monthly figures as they land on my AI job loss statistics page, because one month is weather and 5 in a row is something closer to a pattern.

The health care line is the one I keep returning to. The July report names Montefiore, the Bronx hospital system, alongside the health-data software company Datavant. Montefiore cut 12 nursing positions. In a national table, 12 is a rounding error. In 12 households, it was the whole of that week. It is also 12 people who could tell when a number on a screen did not match the patient in the bed, which is the job every AI deployment depends on and almost never names.

Cut enough of those people and a company ends up renting the model and renting the opinion about whether the model is right. That is the learning debt in its purest form: the retraining a company defers does not disappear, it accrues, and the bill arrives at the next transition when the people who could have judged it are gone. Cutting takes a quarter. Rebuilding that kind of judgment takes a decade. Same line on the same spreadsheet, two different clocks.

Visa reduced its workforce by roughly 7% and attributed the decision to AI-driven efficiency gains. Take the company at its word, since it said so plainly and the efficiency may well be real. The question a 7% reduction does not answer is what Visa will still own in 2031, once the same tools sit in every competitor's stack and the difference between two payment companies comes back down to the people reading the edge cases.

Multiply choices like these across a few thousand companies and they settle questions no shareholder vote ever reaches. Whether the 52-year-old operations manager gets a bridge to her next decade or a severance letter. Whether the cost of this transition sits with companies, with workers, or with the public, which is the substance of the robot-tax argument Bill Gates and Jensen Huang have been having in public. Every monthly report like this one is a few thousand of those answers, already given, before the policy question is anywhere near settled.

If you do not hold the budget, the question still belongs to you. When your employer bought its AI tools, what did it buy for you? A company worth staying at can answer that in a sentence.

I spent 20 years shaping Cisco's $1.1B innovation portfolio. In all of them, the scarce resource was the small number of people who could tell whether a new technology was working, and that number was always smaller than the org chart suggested. So here is the question I am carrying into my own meetings, and you are welcome to it. If every competitor woke up tomorrow with our exact AI stack, what would still be ours? Ask it at your next leadership meeting and watch how long the answer takes. A thin answer is worth discovering in August 2026 rather than in 2031, while there is still time to fund the difference. If you see it differently, I am easy to find.

Here is what makes Alex a credible voice on this topic: Alex spent 20 years shaping Cisco's $1.1B innovation portfolio, deciding which capabilities to build, buy, or cut, which is the same decision every leadership team is making right now under the heading of AI.

Planning a workforce move around AI this year? book a conversation →