Bureau of Labor Statistics

The Bureau of Labor Statistics is the principal US federal agency for labour market measurement, publishing the monthly Employment Situation report. Its March 2026 report recorded +178,000 nonfarm payrolls, beating consensus forecasts of 59,000 and pushing unemployment to 4.3%, even as tech-sector unemployment reached 5.8% — the highest since the dot-com bust. On 14 April 2026 the BLS skipped its scheduled GenAI workplace publication; the New York Fed filled the vacuum the same day with its own Survey of Consumer Expectations data on AI at work.

Established in 1884 within the Department of Labour, the BLS collects data through the Current Population Survey (household) and the Current Employment Statistics programme (employer payroll). Neither survey asks employers to specify whether job eliminations were technology-driven. The gap became politically significant in March 2026 when a bipartisan Coalition of nine senators wrote urging AI-specific displacement tracking. The answer arrived implicitly in April: a Fed Board FEDS Notes paper reconciling three separate BLS-adjacent surveys showed AI adoption rates of 18% (BTOS), 41% (RPS), and 78% (SBU) — a four-fold spread across the same economy — confirming the BLS cannot produce a single authoritative AI adoption figure.

The BLS's limits now structurally define what policy can respond to. The WARN Act captures some mass layoffs but excludes severance-covered workers, contractors, and foreign-routed cuts. The separations data captures when people stop working, not why. Challenger, Gray & Christmas fills the attribution gap voluntarily; Stanford's JOLTS analysis implies even Challenger is a floor, with AI's real labour impact running 34 times Challenger's declared count. That the primary dataset driving the AI displacement debate remains a private outplacement survey is a direct consequence of the BLS measurement gap senators flagged in March.

The BLS has broader macroeconomic significance in the 2026 midterm context through its monthly Employment Situation reports and the Consumer Price Index. Its June 2026 Employment Situation report, released 3 July, showed nonfarm payrolls rising just 57,000, well below the 115,000 consensus forecast; unemployment held flat at 4.2% only because labour-force participation fell to 61.5%, the lowest since March 2021. The soft print landed in the same week as a New York Times poll putting Trump's approval at 39%, reinforcing a summer narrative of economic and political drag arriving together. CPI data from the BLS remains the primary measure against which the impact of tariffs and trade policy is tracked; elevated inflation readings have historically shifted midterm outcomes.

The BLS's GenAI publication, skipped since 14 April 2026 with no public explanation, remains absent ; the New York Fed's Survey of Consumer Expectations continues to serve as the de facto federal measure of AI workplace impact. The BLS operates within the Department of Labour and reports to the Secretary of Labour; its publications are subject to the OMB statistical policy directives that govern federal data releases. The nine-senator letter from March specifically requested AI-specific occupation and displacement codes, which would require a methodological update to the Standard Occupational Classification system, a multi-year process even if authorised today.