On November 5, outplacement firm Challenger, Gray & Christmas published its October report on US job cuts, and the artificial intelligence story inside it cuts against the headlines of the past year. Employers announced 43,994 job cuts in October, up slightly from September but down 71% from October 2025, and the lowest October total since 2023. Of those, only 1,265 were attributed to AI, under 3% of the month and the smallest monthly AI figure of 2026, a year in which AI was cited for as much as 40% of a single month's cuts, in May. Market and economic conditions, not AI, led October with 16,407 cuts, and now total 130,531 for the year, overtaking AI as the leading stated reason for 2026 layoffs. Andy Challenger, the firm's chief revenue officer, described the current environment as a "low-fire" one, with employers holding on to the staff they have heading into the holidays.
That shift sits alongside a second number that points the other way. AI-cited cuts for 2026 stand at 121,401 through October, according to Challenger, more than double the 54,836 the firm attributed to AI across all of 2025. Technology remains the industry hit hardest, with 171,501 cuts so far this year, already above the sector's full 2025 total and 21% higher than the same period last year. In a written statement to HRD America (HCAMag), Elena Magrini, an economist at labor-market analytics firm Lightcast, pushed back on the narrative that AI is replacing workers en masse: "we are not seeing that in the labor data," she said, adding that workplace AI initiatives fail because they start from the wrong question rather than from a business problem AI might help solve.
Why a company's own layoff reason is not an independent measure
Challenger's AI figures come from the reasons employers themselves give in layoff announcements, not from an independent audit of what actually caused each cut. That matters in both directions. A company can cite AI to frame a cost-cutting decision as forward-looking rather than defensive, inflating the AI count; or it can cite restructuring, demand or market conditions when AI tooling was in fact a factor, deflating it. Microsoft's October offer of voluntary exits to 60,000 employees illustrates the lag built into the data: Challenger will not count those departures until workers accept or the company sets a firm target, so a move already underway this quarter may not show up in the numbers for months.
Why it matters even if your company has no plans to cut staff
Most companies reading AI layoff headlines are not deciding whether to cut jobs; they are deciding how much weight to put on AI when they plan hiring, budget for software, or explain a restructuring internally. Both of this month's numbers can be true at once: AI is no longer the reason employers reach for first, and the volume of cuts linked to it this year is still larger than any prior full year on record. Treating either number alone as the whole picture risks either overstating how disposable AI has made a given role, or dismissing a real cost pressure building underneath the slower headline trend.
Four questions before attributing a headcount decision to AI
- Ask what changed in the work, not just in the announcement. If a role is being cut, identify the specific task AI now performs instead of accepting "AI" as a label applied after a cost decision was already made.
- Separate the stated reason from the audited one. Before citing AI internally, in a board update or in a public statement, check whether the claim would survive a line-by-line review of what the software actually replaced.
- Weigh the monthly trend against the yearly total. A single soft month for AI-cited cuts does not undo a year that already outpaces the prior one; plan workforce budgets against the full-year trajectory, not the latest data point.
- Watch for deferred cuts. Voluntary exit programs and unaccepted offers do not show up in job-cut data until later; a quiet quarter can still be followed by a large one once pending programs convert.
On our side, when a client asks whether a role should be cut because "AI can do it now," this is the question we put back to them first: can the company name the specific task being automated, with evidence, or is AI being used as the explanation because it sounds better than the alternative.