28MAR

NY AI layoff law: 162 filings, zero hits

2 min read

19:20UTC

New York required companies to disclose AI's role in mass layoffs. After a year, 162 companies covering 28,300 workers attributed zero cuts to AI.

← Back to The AI jobs data contradicts itself Jump to analysis ↓

EconomicAssessed

Key takeaway

Zero of 162 companies disclosed AI as a factor in layoffs despite a legal obligation to do so.

In 2025, New York State updated its Worker Adjustment and Retraining Notification Act to require companies to disclose AI's role in mass layoffs, becoming the first US jurisdiction to mandate such reporting. After nearly a year of operation, the results are in. ¹ Zero of 162 companies filing layoff notices attributed cuts to AI or technological automation. Those filings covered more than 28,300 workers, including staff at Amazon and Goldman Sachs.

Non-compliance currently carries a penalty of $500 per day. Proposed legislation would raise that to $10,000 per violation and strip companies of state grants and tax incentives for five years. That tougher bill has not advanced.

Silence on this scale is evidence, not absence. Harvard Business Review reported that only 2% of layoffs followed actual AI deployment . Oxford Economics called AI's layoff role "overstated" . Both relied on corporate claims taken at face value. New York's data shows those claims are legally shielded as well as reputationally incentivised. Companies that cut 28,300 jobs had the opportunity and the obligation to say whether AI played a role. Every one said no. Either AI genuinely drives none of the displacement in the nation's financial capital, or the disclosure framework is failing.

Deep Analysis

In plain English

New York passed a law requiring companies to say whether AI played a role when they do mass layoffs. After nearly a year, 162 companies laid off more than 28,000 people, including workers at Amazon and Goldman Sachs. Not one company said AI was involved. The penalty for lying or not disclosing is $500 a day. For billion-dollar companies, that is a trivial fine. Until the penalty is meaningful, there is no incentive to tell the truth.

Deep Analysis

Root Causes

The $500/day penalty is structurally inadequate. For a company like Amazon or Goldman Sachs, potential exposure of $500 per day during a WARN period is a rounding error against litigation risk or reputational exposure from admitting AI-driven displacement. The incentive structure rewards non-disclosure.

Legal uncertainty also suppresses attribution. The definition of AI-driven job loss has not been tested in court. Companies face asymmetric risk: disclosing AI as a reason invites class actions and union bargaining claims, while non-disclosure carries only a civil penalty. Rational legal counsel will advise against attribution until the definition is litigated.

What could happen next?

Consequence
The New York result will be cited in Congressional debates as evidence that voluntary disclosure frameworks cannot generate honest AI attribution data, strengthening the case for mandatory federal reporting with meaningful penalties.
Short term · High
Risk
Other states considering WARN Act amendments may model weak penalty structures on New York, producing the same zero-attribution outcome and wasting a decade of potential evidence collection.
Medium term · Medium
Precedent
New York's failure is the most important data point in the AI disclosure debate: it proves empirically that disclosure laws without credible enforcement produce no data.
Long term · High

Source Landscape

This story draws on mixed-leaning sources from United States

United States

Primary parallel: The original 1988 federal WARN Act required 60-day advance notice for mass layoffs. Studies by the GAO in 1993 found systematic under-compliance in the first five years, with companies structuring cuts in batches below the threshold and citing business exceptions to avoid filing. Compliance improved only after states introduced their own higher-penalty versions.

Counter-parallel: California's WARN Act amendments of 2003, which extended protections and increased penalties, produced substantially higher compliance and generated a usable dataset on mass layoffs within two years. The model shows disclosure laws can work when enforcement is credible.

Consensus view: Employment law scholars at Cornell's ILR School note that the New York result mirrors the experience of the original 1988 federal WARN Act, which produced systematic under-reporting for its first three years because penalties were too low to deter non-compliance and enforcement required employee-initiated complaints.

Counter-view: Business lobby groups including the Partnership for New York City argue the zero attribution reflects genuine absence of AI-driven cuts rather than disclosure avoidance, citing Oxford Economics' finding that AI's layoff role is overstated.

Key tension: Whether zero attributions under an unenforced $500/day penalty structure proves avoidance or proves absence.

Sources:Bloomberg Law·Bloomberg Law

Mentions:Amazon →Goldman Sachs →Oxford Economics →Prima →New York State →Harvard Business Review →

First Reported In

Update #3 · The AI jobs data contradicts itself

Bloomberg Law· 28 Mar 2026

Read original →

Causes and effects

This Event

NY AI layoff law: 162 filings, zero hits

The failure of the first AI layoff disclosure law proves that voluntary compliance frameworks cannot generate honest data on displacement.

Led to

Oracle India: 12,000 cut by email

New York WARN Act failure to capture any AI attributions is compounded by Oracle India cuts falling entirely outside US measurement systems.

Occurred 31 Mar 2026

Read story →

Three in four never file claims

New York WARN Act failure to capture AI attributions extends the measurement failure documented when 75% of unemployed never file.

Occurred 28 Mar 2026

Read story →

Different Perspectives

Barclays

Barclays economist Pooja Sriram flagged a 28,000-a-month bleed in finance and information roles the same week Microsoft disputed that AI drove its own 4,800 cuts. The bank treats Challenger's AI-attribution share as a lagging indicator against faster erosion visible in raw labour-market data.

European Commission

Brussels deferred the Digital Omnibus's Annex III employment-compliance deadline from 2 August 2026 to December 2027, even as California advanced three binding AI-hiring bills the same week. The 17-month delay leaves EU workers without the algorithmic-hiring safeguards the regulation already promises.

OpenAI

OpenAI proposed a 5% US government equity stake worth $42.6bn, structured as a public wealth fund modelled on the Alaska Permanent Fund, with Sam Altman pitching it directly to Trump, Bessent and Lutnick. The offer pre-empts Sanders' rival one-time 50% AI-stock tax, which has not yet reached committee.

India's IT and outsourcing sector

BAT's transfer of 3,500 roles to Accenture on 29 June fits a delivery model Indian IT firms increasingly run: consultancies win Western contracts, then execute through offshore centres. The sector expects more Fit2Win-style transfers, not straight redundancies, as employers absorb AI without cutting outsourced headcount.

European Trade Union Confederation

ETUC says the Council's shift from 'ensure' to 'support' in the AI-literacy duty, confirmed in the Digital Omnibus's final adoption on 29 June, is a collapse of the legal threshold, not a drafting tidy-up. It expects EU workers to face AI-driven hiring and monitoring decisions with a statutory right to explanation that exists in name only.

British American Tobacco's Fit2Win workforce

BAT is cutting 9,000 roles under Fit2Win, transferring 3,500 to Accenture rather than making them redundant, to reach roughly £500m in AI-driven savings by 2027. For affected staff, that distinction decides whether they keep a job at all, just not at BAT.