8JUN

US workers use AI more than Europeans

3 min read

11:04UTC

An NBER study found 43% of US workers use generative AI at work against 32% in six European countries, and pinned the gap on management practice rather than regulation.

← Back to Jobs report says fine, layoff report says no Jump to analysis ↓

EconomicDeveloping

Key takeaway

US-Europe AI adoption differs by management practice, not regulation, an NBER study found.

The National Bureau of Economic Research (NBER), a US non-profit that publishes economic working papers, found that 43% of US workers use generative AI at work against 32% across six European countries surveyed in early 2026 ¹. The paper, titled "Mind the Gap", attributed most of the 11-point difference to management practices rather than regulatory differences.

The finding cuts against a common framing. European AI rules, including the workplace provisions of the EU AI Act that Brussels delayed to December 2027 , are often blamed for slower uptake on the continent. NBER's data points instead at how firms organise work: whether managers deploy the tools, train staff on them and rebuild processes around them. The paper's authors conclude the lighter-regulation argument for faster adoption explains little of the 11-point gap.

The distinction matters for the displacement debate. If adoption is driven by management choices, the pace at which AI reaches the workplace, and the jobs it reshapes, is set largely inside firms rather than by lawmakers. That puts more of the labour-market outcome in the hands of company decisions than of the regulatory contest now playing out across US states and Brussels.

Deep Analysis

In plain English

A study from the National Bureau of Economic Research (a private US economics research organisation) published in June 2026 found that 43% of US workers say they use AI tools in their job, compared with 32% of workers in six European countries surveyed at the same time. The researchers traced the gap to management practices rather than laws or regulations: US companies measure individual output more closely and push employees harder to adopt new tools. European workplaces tend to require worker consultation before rolling out new technology. European workplaces tend to have more worker consultation before rolling out new technology. The finding matters because higher AI adoption typically precedes a restructuring decision: once workers are using AI tools widely, managers begin asking whether they need as many workers to do the same volume of work.

What could happen next?

Consequence
If the NBER management-practice explanation is correct, European AI adoption will accelerate as firms adopt US performance management norms under investor pressure, bringing the European displacement curve two to four years behind the US rather than avoiding it.

Source Landscape

This story draws on neutral-leaning sources from United States

United States×2

LeftRight

Primary parallel: The 1990s US-Europe personal computer adoption gap showed the same management-practice attribution: US firms adopted PCs faster because performance management systems tied output measurement to individual machine use, while European firms with stronger job security norms and union consultation requirements adopted more slowly.

The gap narrowed from roughly 15 percentage points in 1993 to near parity by 2001 as European regulatory frameworks adapted. The NBER w34995 finding replicates that 1990s pattern with an almost identical adoption gap size.

Counter-parallel: The 2000s internet adoption gap did NOT close uniformly: Scandinavian countries exceeded US adoption rates by 2007 because state broadband investment and flat organisational cultures accelerated uptake beyond the management-practice advantage. If EU AI Act obligations create structured adoption incentives alongside monitoring, the European trajectory could diverge from the 1990s PC pattern.

Consensus view: NBER working paper w34995 and the MIT Sloan Management Review's research on management technology adoption both identify US management practices, particularly performance management systems that reward measurable individual output, as the primary driver of faster AI tool adoption.

US knowledge workers face more granular productivity monitoring than their European peers, creating stronger individual incentives to adopt tools that visibly increase throughput. The 43% versus 32% gap persists even controlling for industry composition and regulatory environment.

Counter-view: The Hans Boeckler Foundation (Germany) and France Strategie both argue the NBER's 'management practice' explanation conflates cause and effect. European firms show lower adoption in part because works council consultation requirements add 4-8 weeks to new tool deployment decisions.

The NBER paper controls for sector but not for mandatory co-determination processes, which apply to 40% of the European workers surveyed. The gap may partly reflect structural governance differences rather than cultural management style.

Key tension: Whether the US-Europe adoption gap will persist as a durable divergence (because management practice cultures are sticky) or narrow as European co-determination processes adapt AI consultation procedures and the EU AI Act's workplace obligations come into force.

Sources:Bureau of Labor Statistics·Challenger, Gray & Christmas·Cambridge Centre for Alternative Finance

Mentions:EU AI Act →Brussels →Germany →France →National Bureau of Economic Research →

First Reported In

Update #12 · Jobs report says fine, layoff report says no

Bureau of Labor Statistics· 8 Jun 2026

Read original →

Different Perspectives

European workers and regulators

NBER working paper w34995 found European workers use generative AI at 32% versus 43% of US workers, a gap driven by management practice rather than regulation. The EU AI Act's high-risk employment deadline stays at December 2027, leaving European workers facing the same displacement curve two to four years behind the US.

AI industry (Leading the Future PAC, OpenAI, Andreessen Horowitz)

Leading the Future committed over $100 million to the 2026 midterms and targeted regulation-minded candidates in the 2 June primaries; its counter-fund Public First formed at $50 million. The PAC runs advertising on healthcare and jobs without naming AI, mirroring the 1994 insurance industry campaign that defeated the Clinton health plan.

UK youth entering the labour market

UK youth unemployment reached 14.7% in January-March 2026, the highest since 2014, with 22.7% of young jobseekers out of work more than a year. The ONS publishes no AI-exposure breakdown, so policy is being set blind to the channel doing the damage.

US displaced workers (tech and finance)

Tech workers face median reemployment times of 4.7 months, up 47% from 2024, with a hiring pool contracting faster than AI-specialist openings can absorb them. Finance operations workers are the next cohort: 52% of their employers now run agentic AI in the exact functions where most of them work.

TSMC and Taiwan chip supply chain

Nvidia's 17% headcount growth to 42,000 on $81.6 billion in quarterly revenue depends on TSMC's CoWoS advanced packaging capacity constraining H100 and B200 supply, sustaining margins above 70%. The AI build-out's sole headcount-growth story runs through a Taiwan supply chain that has no parallel in downstream software.

Displaced tech workers globally

CrowdStrike's SEC disclosure puts AI attribution on a material regulatory record for the first time, but Oracle's Massachusetts WARN clock expired unfiled after up to 14 workers were logged as remote despite office proximity. The legal apparatus cannot enforce what it cannot see: hybrid reclassification, GCC transfers, and hires never made.