Skip to content
You can now search across every topic, entity and event.What's new
AI: Jobs, Power & Money
27JUL

CFOs see AI job cuts nine times higher

3 min read
10:02UTC

Seven hundred and fifty chief financial officers told the Atlanta Fed they expect AI-attributed layoffs to be nine times higher in 2026. The number is smaller than it sounds.

← Back to SAP freezes R&D headcount as others deny AIJump to analysis ↓
EconomicDeveloping
Key takeaway

AI job cuts are projected to rise ninefold but affect only 0.4% of the US workforce.

The Federal Reserve Bank of Atlanta and the National Bureau of Economic Research published a survey of 750 US chief financial officers on 25 March, finding that projected AI-attributed job cuts for 2026 are nine times higher than 2025 levels. 1 The raw number is roughly 502,000 roles. As a share of the total US workforce, that is 0.4%.

The figure is the first hard employer-side estimate of AI displacement at scale. It arrives in a quarter where Q1 tech layoffs had already reached 59,000 , up from 45,363 at the last count, with one in five cuts explicitly citing AI. The ninefold increase sounds severe. The denominator tells a quieter story: 0.4% falls within normal labour market churn, where monthly separations run at roughly 3.5%.

Only 44% of surveyed CFOs plan AI-related layoffs at all. 2 The majority intend no cuts. The survey also documents a productivity paradox: executives perceive AI gains that do not yet appear in revenue. Companies are cutting based on expected capability, not demonstrated return on investment. That pattern produced the 55% regret rate Orgvue found earlier this quarter , when more than half of leaders who cut staff for AI admitted they were wrong.

The Atlanta Fed data does not contradict Harvard Business Review's finding that only 2% of layoffs followed actual AI deployment . It refines it. Firms intend to cut. They have not done so at scale. When they do, the numbers will be smaller than the headlines suggest.

Deep Analysis

In plain English

A US Federal Reserve survey asked 750 finance directors at large companies how many jobs they plan to cut because of AI this year. Their answer: about nine times as many as last year, roughly 500,000 positions across the whole country. That sounds huge. But the US has about 160 million workers. Five hundred thousand is less than one half of one percent. Normal workplace turnover each month accounts for more than that. The alarming headline and the modest denominator are both true at the same time.

Deep Analysis
Root Causes

Corporate earnings pressure from rising AI infrastructure costs creates an incentive to signal efficiency to equity markets. Block's 22-25% share-price surge after a 40% workforce cut established a template: headcount reduction plus AI investment narrative produces immediate shareholder reward, regardless of actual productivity gain.

The productivity measurement gap is structural. Firms perceive AI gains that do not yet appear in revenue, so they project future cuts based on anticipated capability rather than demonstrated return. This inflates CFO forecasts above what deployment data would support.

Fiscal drag compounds the effect. With 75% of US federal tax revenue derived from labour taxation, even a 0.4% displacement concentrated in high-income roles disproportionately erodes the tax base relative to its share of the workforce.

What could happen next?
  • Consequence

    CFO intent will translate into announced layoffs, pushing Q2 2026 tech job cuts well above the 59,000 recorded in Q1.

    Short term · Medium
  • Risk

    If the 44% of CFOs planning cuts act in the same quarter, the concentrated impact on specific sectors could trigger a Sahm Rule indicator signal despite low headline unemployment.

    Short term · Low
  • Opportunity

    The gap between projected cuts (502,000) and the 1.6 million unfilled AI roles creates a retraining arbitrage window for workers who can pivot to AI-adjacent skills before the labour market tightens.

    Medium term · Medium
  • Precedent

    This is the first Federal Reserve paper to quantify forward AI displacement from the employer side, establishing a baseline against which future surveys will be measured.

    Long term · High
Sources:Federal Reserve Bank of Atlanta / NBER·Federal Reserve Bank of Atlanta / NBER
Mentions:Oxford Economics →Prima →Orgvue →Claudia Sahm →Federal Reserve →Harvard Business Review →National Bureau of Economic Research →
First Reported In

Update #3 · The AI jobs data contradicts itself

Federal Reserve Bank of Atlanta / NBER· 28 Mar 2026
Read original →
Causes and effects
This Event
CFOs see AI job cuts nine times higher
The first large-scale employer survey to quantify 2026 AI displacement projects 502,000 cuts, reframing the debate from panic to proportion.
Led to
AI now leads all US layoff reasons
Atlanta Fed 502,000 projection framed the scale against which March 15,341 figure lands.
Occurred 1 Apr 2026
Read story →
Oracle cuts 30,000 to fund AI
Atlanta Fed 502,000 projection anticipated the intention-to-action closure that Oracle 31 March action represents.
Occurred 31 Mar 2026
Read story →
Two thirds of CEOs freeze hiring
Atlanta Fed 502,000 projection aligns with 66% of CEOs planning freezes; both indicate sustained corporate withdrawal from hiring.
Occurred 29 Mar 2026
Read story →
Meta codes its own org chart
Atlanta Fed CFO survey projected AI-attributed cuts nine times 2025 levels; Meta's 8,000-cut restructuring is the first single-firm realisation of that projection at scale.
Occurred 19 Apr 2026
Read story →
Goldman counts 25,000 jobs lost monthly
Goldman's 25,000/month substitution model provides a bottom-up mechanism explaining the Atlanta Fed CFO projection of 502,000 annual cuts.
Occurred 6 Apr 2026
Read story →
Different Perspectives
European Commission
European Commission
The European Commission's draft Annex III guidelines, closed for comment on 23 July, treat algorithmic scoring in recruitment, pay and termination as high-risk regardless of whether a human signs off, echoing Spain's Audiencia Nacional ruling 101/2026 on concealed scheduling algorithms. Brussels is shifting the fight from counting AI job losses to assigning legal liability for the tools themselves.
Office for National Statistics
Office for National Statistics
The Office for National Statistics recorded UK vacancies rising to 712,000 on 21 July, the first quarterly increase this beat has tracked, with payrolled employment down 85,000 on the year against May's 210,000 fall. The bulletin names no AI cause anywhere, and that is the point: nothing in the release confirms the displacement story it gets cited to support.
Christian Klein, SAP
Christian Klein, SAP
Christian Klein told investors on 23 July that SAP's research headcount will not grow for twelve months because AI agents and their token costs are absorbing the work, not because SAP is cutting jobs. He frames it as commercial arithmetic: the cost of AI-assisted coding tokens plus the salaries specialist AI hires command, not people being replaced by machines.
Betsey Stevenson, University of Michigan
Betsey Stevenson, University of Michigan
Betsey Stevenson argued that the 187,000 jobless-claims reading describes a market that hires little and fires little, not one AI is emptying. She said the real damage hides in eligibility rules and suppressed job postings, not in the headline layoff counts employers keep denying.
Comisiones Obreras, UGT and Concentrix's A Coruña works committee
Comisiones Obreras, UGT and Concentrix's A Coruña works committee
Comisiones Obreras, UGT and Concentrix's A Coruña works committee blamed Microsoft's push toward AI self-service for the 80 redundancies unions signed off on 22 July, not unavoidable business cause. A second Coruña procedure covering 80 more jobs runs to a 31 August deadline, and the unions want the state, not the employer, setting the pace of AI-driven cuts.
Stanford's 'We Must Act Now' signatories
Stanford's 'We Must Act Now' signatories
More than 200 academics, including 16 Nobel laureates, published a 13 July letter warning of AI-driven labour disruption, citing Daron Acemoglu's NBER estimate that AI's total factor productivity gain stays under 0.66% over ten years. The letter's own cited economics sit well below Goldman Sachs Research's 1.5-percentage-point estimate published the same week.