Everything About AI
NextFin News - This earnings season has delivered a verdict on artificial-intelligence job cuts that is anything but uniform: companies that frame headcount reductions as the funding source for a credible AI buildout are being rewarded, while those cutting to defend margins beneath a swelling capital-spending bill are being punished. The split shows a market that has stopped applauding layoffs for their own sake and started grading every reduction on what it buys.
The Situation: Cuts Are Back, With Cost and AI Sharing the Blame
Mentions of job cuts, firings and layoffs are trending again in news coverage, with October references the highest since May, according to a news analysis. The official tally is starker. U.S. employers announced 153,074 job cuts in October, a 175% jump from a year earlier and the highest October total since 2003, Challenger, Gray & Christmas reported. Through October, employers have announced 488,077 cuts for the year.
The reason companies give matters, and October showed two drivers sharing the load. Among private-sector employers, cost-cutting was the top-cited reason, responsible for 50,437 announced layoffs. Artificial intelligence was second, with 31,039 cuts as companies restructure and automate. AI has been cited for 48,414 cuts so far this year. That follows a different pattern earlier in 2026: through July, AI led the reason rankings for five consecutive months, cited in 112,713 cuts, roughly 24% of the total. Since the outplacement firm began tracking AI as a distinct category in 2023, it has been linked to 184,538 announced cuts.
Technology remains the center of gravity. The sector announced 33,281 cuts in October alone, up sharply from 5,639 in September, and 141,159 for the year, a 17% increase. "Tech remains the center of gravity for this year's cuts, and AI is still the reason companies give," said Andy Challenger, chief revenue officer at Challenger, Gray & Christmas.
But the cuts have spread well beyond software. Banks, insurers, chemical producers, logistics companies and media firms have all cited AI or automation in 2026 restructuring plans. Allianz is cutting up to 1,800 roles in its travel insurance division as AI replaces manual work. SEB is eliminating up to 2,100 positions by the end of 2027 in a restructuring designed to leverage AI. Wisetech Global cut 2,000 jobs, one-third of its workforce, citing AI integration. American Express said in October it is eliminating 5% of its global workforce, roughly 3,850 roles from a base of almost 77,000 employees at the end of 2025, while taking a $336 million restructuring charge. Intuit is cutting roughly 3,000 roles with an increased focus on AI efforts. FICO is cutting about 15% of its workforce, roughly 570 roles, to simplify its operating structure around AI workflows. DNB is cutting roughly 400 roles because of increased AI investment. In Europe, governments are adding to the number of jobs at risk as public employers pursue their own automation agendas.
The counterpoint is that the broader labor market has not broken. Through July, employers announced 477,033 job cuts, down 41% from the first seven months of 2025, and hiring plans rose 25% over the prior year.
"Hiring has also increased over last year by 25%, so while AI is shifting the labor market, it is not dismantling it."
That is Andy Challenger, chief revenue officer at Challenger, Gray & Christmas. The unemployment rate stood at 4.2% in September, and openings per unemployed worker stood at 1.0, below the 2019 level.
The Split: What the Cuts Buy
The market is not reacting to the layoffs themselves. It is reacting to the balance sheet behind them.
On one side of the split are the cuts that buy credibility. When a company pares payroll and channels the savings into a visible, revenue-producing AI buildout, investors treat the reduction as capital reallocation rather than retrenchment. JPMorgan Chase will boost its annual technology spending by nearly $2 billion to $19.8 billion in 2026, driven largely by AI initiatives, according to the bank's finance chief. Standard Chartered's shares reached a new high after its automation-led restructuring, which CEO Bill Winters described as replacing lower-value human capital with financial and investment capital.
On the other side are cuts that merely offset a capital bill investors no longer trust. The spending commitments are enormous: one research house raised its aggregate AI capital-expenditure forecast to $1.0 trillion in 2026, $1.4 trillion in 2027 and $1.5 trillion in 2028, supported by stronger hyperscaler commitments. The praise came with immediate discipline. Alphabet maintained its $195 billion to $205 billion 2026 capital plan but reported negative free cash flow and signaled a further increase in 2027; its shares fell about 18% over the 30 days after the report. Meta lifted the lower end of its 2026 capital range to $135 billion to $145 billion and posted negative third-quarter free cash flow; its shares fell nearly 7% the following day.
The mechanism is straightforward: labor expense is being converted into computing capacity. Meta recorded $1.2 billion of severance in the second quarter for a reduction affecting about 8,000 employees. That same quarter, capital expenditures ran $31.1 billion, against operating cash flow of $31.9 billion. The severance savings are small next to the infrastructure tab. The cuts are a byproduct of capital allocation, not a substitute for it. Investors who follow that math reward the companies where AI revenue is visible; they punish the ones where spending commitments outrun monetization.
The Two Meanings of "Because of AI"
A second source of investor division is ambiguity. "Because of AI" can mean two different things: software now handles the work, or management wants the payroll money for computing power. The distinction matters because the first is a productivity story and the second is a funding story, and they carry different risks.
Cloudflare, cutting more than 1,100 employees, framed its move as a structural reorganization for the "agentic AI era" rather than a cost-cutting exercise, in an email from CEO Matthew Prince and President Michelle Zatlyn. Wix said it was cutting around 20% of staff, about 1,000 roles, because "we need to become a faster, leaner, and flatter organization," according to co-founder and CEO Avishai Abrahami. Uber is cutting 3,300 jobs, roughly 10% of its workforce, to remove layers and simplify team structures, CEO Dara Khosrowshahi said.
American Express offered the cleanest articulation of the sequencing argument. The restructuring is meant to make the company "leaner, more agile, and also to prepare for AI," CFO Christophe Le Caillec said.
"You don't want to automate processes that are not optimally structured, and so what we are doing here is to reduce layers and increase spans of controls, so that we can automate and take advantage of AI in the coming quarters and the coming years."
That sequencing is the pitch investors want to hear: cut first, automate second, grow margins third. The risk is that the middle step never arrives. If the AI-driven productivity gain does not materialize, the company is left with a smaller organization carrying a larger fixed cost in depreciation and cloud commitments. That is the trade investors are now pricing, and it is why two companies can announce similar headcount reductions and move in opposite directions.
The Second-Order Effect: Accounting Follows the Capital
The capital-for-labor swap is large enough to change corporate accounting, and that is where the second-order risk hides. As AI servers become the dominant asset on technology balance sheets, companies are stretching depreciation schedules to soften the earnings hit. One company extended its AI server depreciation life to 5.5 years, up from four-year lives earlier in the decade. The move does not change the cash spent; it changes when the expense reaches the income statement.
That accounting choice matters for the investor split. A longer depreciation life makes near-term earnings look stronger while pushing the true economic cost further into the future. If AI revenue arrives on schedule, the extended schedule is harmless. If monetization lags, the company has merely deferred the recognition of a cost that is already incurred, and the eventual earnings catch-up will be larger. Investors grading the quality of a cut are therefore also grading the accounting that comes with the capital plan.
The financial sector shows the same pattern at scale. In the first quarter of 2026, the six largest Wall Street banks collectively cut 15,000 jobs while posting roughly $47 billion in combined profit, an 18% year-on-year increase. The trade is explicit: substitute financial capital for human capital, and let the machines carry the marginal work. The question for shareholders is identical to the one technology investors are asking: does the revenue arrive before the depreciation does?
Cyclical or Structural: The Call
This is a structural shift in how work is composed and how capital is allocated, layered on top of a cyclical labor market that remains intact. The evidence for the structural leg is the permanence of the capital commitment: $1 trillion in AI spending is not a one-quarter adjustment; it is a multi-year reweighting of corporate balance sheets away from labor and toward computing. Once that capital is deployed and those tasks are re-engineered, the change does not reverse on its own.
The evidence for the cyclical leg is the low-fire labor market itself. The unemployment rate was 4.2% in September, the layoff rate remains close to record lows, and the quits rate is depressed, meaning few workers are moving voluntarily. New postings on job boards have yet to improve meaningfully, and posted wage growth ran 2.4% year over year against consumer-price inflation of 3.7%. This is the profile of a market where losing a job is rare but replacing one is slow: the average search runs 24.8 weeks, 27.1% of the unemployed have been out of work for 27 weeks or more, and recent college graduates face an unemployment rate of about 5.6% with 42% in jobs that do not require a degree.
Separating the two matters because they imply different endings. If the phenomenon were purely cyclical, the cuts would reverse as demand recovered. They will not, because the capital has already been committed. If it were purely structural, the unemployment rate would already be rising sharply. It is not, because the cuts are concentrated in specific functions and companies rather than spread across the economy. The net effect is a labor market that is stable in aggregate but fragile at the margin, adjusting function by function rather than through a single aggregate shock.
The Counter-Thesis
The strongest case against this reading is that investors are not split at all: they simply love cost cuts, as they always have, and the AI narrative is decoration. Cutting labor costs signals to shareholders that executives are serious about margins, and stocks have historically risen on layoff announcements for that reason alone. On this view, the punishment of heavy capital spenders is a separate trade, not evidence of a new grading rubric.
That argument has force, but it does not survive the cross-section. If investors were rewarding cuts indiscriminately, reductions across the same cohort of large technology and financial companies would have produced uniform gains. They did not. The market differentiated within the group, rewarding the names where cloud backlog growth and AI revenue are most visible and punishing the ones where spending commitments outran monetization. The split is inside the layoff cohort, which is exactly what a funding-quality thesis predicts and what a generic cost-cut thesis cannot explain.
The signal that would prove the funding-quality reading wrong is a sustained rise in the overall layoff rate above roughly 1.5% of employment alongside an unemployment rate climbing past 4.5% while the hiring rate stays depressed. That combination would mean the cuts have escaped the AI-and-efficiency cohort and become broad-based demand destruction. At that point the market would stop grading cuts on quality and start pricing a recession.
What's Next
Short term, the split should persist through the remainder of the earnings season. Companies that pair headcount reductions with concrete AI revenue figures and free-cash-flow discipline are likely to be rewarded; those that announce cuts alongside rising capital guidance without a monetization path face the same punishment the market delivered in October.
Medium term, the question is whether the productivity gains arrive. If AI-driven efficiency shows up in margins without a commensurate rise in capital intensity, approval will broaden beyond the hyperscalers to the banks, insurers and industrial companies now restructuring around automation. If margins improve only because headcount fell while depreciation climbed, multiples will compress across the cohort.
Long term, the structural leg dominates. Corporate balance sheets are being reweighted toward computing and away from labor for the rest of the decade. The winners are the companies that convert payroll savings into revenue-producing capacity; the exposed are the ones that use AI as a margin-defense story while their fixed-cost base swells.
The base case is a continued split: cost discipline earns a pass, but only when paired with a credible path to AI revenue. The upside case is that productivity arrives faster than expected, lifting margins and wages together and letting the labor market absorb the displaced. The downside case is that capital spending keeps climbing while returns lag, forcing a second, deeper wave of cuts that finally tips the low-fire labor market into something colder.
The market is no longer asking whether a company is cutting jobs. It is asking what the company is buying with the cuts, and whether it can afford the bill.
Explore more exclusive insights at nextfin.ai.
