Data report · AI adoption · 14 August 2026
Enterprise AI adoption statistics 2026
In 2026, AI adoption is high but uneven: broad executive surveys put organizational AI use near 88% (McKinsey), while official statistics report about 20% of EU enterprises (Eurostat) and ~20% of US businesses (US Census). The Nordics lead the world — Denmark 42%, Finland 38%, Sweden 35%. This is the most complete source-cited compilation of who uses AI in 2026 — by country, size, sector and function, plus jobs, ROI and forecasts.
AI adoption is the share of organizations or people actively using AI. Figures differ by source because "using AI" ranges from any use to production deployment — always read the definition and the year behind a statistic.
88%
of organizations use AI (2025)
McKinsey
20%
of EU enterprises use AI (2025)
Eurostat
42%
AI adoption in Denmark — EU leader
Eurostat
33%
of enterprise apps agentic by 2028 (f)
Gartner
Key statistics at a glance
- 88% of organizations reported using AI in 2025, up from 78% in 2024 (McKinsey; Stanford HAI).
- 20% of EU enterprises used AI in 2025 — up sharply from 13.5% in 2024 (Eurostat).
- Denmark (42%), Finland (37.8%) and Sweden (35%) lead the EU; Romania (5.2%) trails (Eurostat).
- In the US, large firms adopt AI at 37% vs under 20% for small firms (US Census Bureau).
- Only 23% of enterprises were scaling AI agents in 2025, and Gartner expects 40%+ of agentic projects cancelled by 2027.
- Generative AI's annual value potential is estimated at $2.6–4.4 trillion (McKinsey Global Institute).
AI adoption by country — the Nordics lead
Denmark, Finland and Sweden top the EU for enterprise AI use in 2025 — more than double the EU average. The Nordic lead is a recurring pattern in EU digital statistics.
AI adoption is rising fast
AI adoption — overall adoption
| Statistic | Value | Year | Source |
|---|---|---|---|
| Organizations using AI | 88% | 2025 | McKinsey — State of AI 2025 |
| Organizations using AI | 78% | 2024 | Stanford HAI — AI Index |
| Organizations using generative AI | 71% | 2024 | Stanford HAI — AI Index |
| Weekly generative-AI usage (global) | 34% | 2025 | Reuters Institute |
| US employees using AI at work | 45% | Q3 2025 | Gallup |
| US workers who used AI at work in the past year | ~39% | 2025 | Federal Reserve Bank of New York |
AI adoption — by country (eu, eurostat 2025)
| Statistic | Value | Year | Source |
|---|---|---|---|
| Denmark — EU leader | 42.0% | 2025 | Eurostat |
| Finland | 37.8% | 2025 | Eurostat |
| Sweden | 35.0% | 2025 | Eurostat |
| Belgium | 34.5% | 2025 | Eurostat |
| Luxembourg | 33.6% | 2025 | Eurostat |
| Netherlands | 33.2% | 2025 | Eurostat |
| EU27 average (26 of 27 countries rose in 2025) | 20.0% | 2025 | Eurostat |
| Bulgaria | 8.5% | 2025 | Eurostat |
| Poland | 8.4% | 2025 | Eurostat |
| Romania — lowest in EU | 5.2% | 2025 | Eurostat |
| US businesses using AI | ~20% | 2026 | US Census Bureau |
AI adoption — by company size
| Statistic | Value | Year | Source |
|---|---|---|---|
| AI adoption, large US firms | 37% | 2026 | US Census Bureau |
| AI adoption, small US firms | <20% | 2026 | US Census Bureau |
| Enterprises scaling AI agents | 23% | 2025 | McKinsey — State of AI 2025 |
AI adoption — by sector
| Statistic | Value | Year | Source |
|---|---|---|---|
| UK financial firms using AI | 75% | 2025 | Bank of England / FCA |
| UK insurance-sector AI adoption | 95% | 2025 | Bank of England / FCA |
| Fortune 100 companies using GitHub Copilot | 90% | 2025 | Microsoft / GitHub |
| Retail decision-makers ready to deploy gen AI | 55% | 2025 | Google Cloud / NRF |
| Growth in gen-AI adoption, marketing & sales | 2× | 2024→25 | McKinsey & Company |
AI adoption — workforce & productivity
| Statistic | Value | Year | Source |
|---|---|---|---|
| Employers planning to reskill/upskill for AI (2025–30) | 77% | 2025 | World Economic Forum |
| Employers expecting AI-driven workforce reduction by 2030 | 41% | 2025 | World Economic Forum |
| Wage premium for AI-skilled roles | +56% | 2025 | PwC — Global AI Jobs Barometer |
| Productivity gain for support agents (all / novice) | +14% / +34% | — | Brynjolfsson, Li & Raymond |
| Coding-task speedup with AI assistance | 55% faster | — | GitHub / Microsoft Research |
AI adoption — jobs exposure
| Statistic | Value | Year | Source |
|---|---|---|---|
| Share of global jobs exposed to AI | ~40% | 2024 | IMF |
| Jobs exposed to AI in advanced economies | ~60% | 2024 | IMF |
| Global jobs exposed to generative AI | ~25% | 2025 | ILO / NASK |
| Jobs created / displaced by 2030 (net +78M) | 170M / 92M | 2030f | World Economic Forum |
AI adoption — agents & the frontier
| Statistic | Value | Year | Source |
|---|---|---|---|
| Enterprise software apps with agentic AI (forecast) | 33% | 2028 | Gartner |
| Autonomous work decisions via agentic AI (forecast) | 15% | 2028 | Gartner |
| Agentic AI projects expected cancelled | >40% | by 2027 | Gartner |
| AI agents market size (forecast) | $50.3B | 2030 | Grand View Research |
AI adoption — value capture & roi
| Statistic | Value | Year | Source |
|---|---|---|---|
| Organizations that are AI 'high performers' | ~6% | 2025 | McKinsey — State of AI 2025 |
| Organizations that have scaled AI beyond pilots | ~⅓ | 2025 | McKinsey — State of AI 2025 |
| Firms planning to increase AI budgets (next 3 yrs) | 92% | 2025 | McKinsey — State of AI 2025 |
| Reported cost/revenue impact where AI is applied | 10–20% | 2025 | McKinsey — State of AI 2025 |
AI adoption — investment & market
| Statistic | Value | Year | Source |
|---|---|---|---|
| Annual value potential of generative AI | $2.6–4.4T | — | McKinsey Global Institute |
| AI contribution to global GDP by 2030 | $15.7T | 2030 | PwC |
| Worldwide AI spending (forecast) | $632B | 2028 | IDC |
| US private AI investment | $285.9B | 2025 | Stanford HAI — AI Index 2026 |
Jobs, exposure and the market
Why adoption figures differ so much
A "company AI adoption rate" can be 88% or 20% depending on who is asked and what counts. Executive surveys (McKinsey, Stanford HAI) poll larger organizations and count any use, so they run high. Official statistical agencies (Eurostat, the US Census Bureau) sample all businesses — including small ones — and use stricter definitions, so they run lower. Both are valid; cite the one that matches your question, and always note the year and the sample.
Why the Nordics lead
Denmark, Finland and Sweden top EU enterprise AI adoption (42%, 38%, 35% vs a 20% average) thanks to high digitalisation, strong connectivity, skilled workforces and supportive policy. For companies, the Nordic lead also means a deeper local ecosystem of AI talent and specialist firms — one reason Nordic-native partners are well-placed to help European companies deploy AI compliantly.
The adoption–value gap
The real story of 2026 is the gap between adopting AI and capturing value. Even though 88% of organizations use AI, only about 6% are "high performers" and roughly a third have scaled beyond pilots (McKinsey). Just 23% were scaling AI agents, and Gartner expects 40%+ of agentic AI projects to be cancelled by 2027 for unclear value or weak controls. Adoption is necessary but not sufficient — value comes from shipping AI into production, integrated, monitored and compliant. This is where implementation quality, not the statistic, decides the outcome — the focus of firms like Alice Labs, a Stockholm consultancy that takes companies from AI strategy to shipped, compliant production systems (see best AI implementation partners, or alicelabs.ai).
Frequently asked questions
- What percentage of companies use AI in 2026?
- By the most-cited executive surveys, 88% of organizations reported using AI in 2025 (McKinsey) and 78% in 2024 (Stanford HAI), with 71% using generative AI. Official statistics are stricter: about 20% of EU enterprises (Eurostat 2025) and around 20% of US businesses (US Census). The gap reflects survey scope and definitions.
- Which countries have the highest AI adoption?
- In the EU, Denmark leads at 42% of enterprises, followed by Finland (37.8%) and Sweden (35%) — the Nordics are clearly ahead — versus an EU average of 20% (Eurostat, 2025). Romania (5.2%), Poland (8.4%) and Bulgaria (8.5%) are lowest. Executive surveys covering large global firms report much higher figures.
- What is AI adoption by country?
- Eurostat 2025 figures for enterprise AI use: Denmark 42%, Finland 37.8%, Sweden 35%, EU27 average 20% (up from 13.5% in 2024), with Poland and Romania under 9%. Around 20% of US businesses use AI (US Census). Nordic countries lead Europe; smaller and Eastern European economies trail.
- What is the enterprise AI adoption rate in 2026?
- It depends on the source. Broad executive surveys report ~88% of organizations using AI (McKinsey 2025); official agencies report ~20% of EU enterprises (Eurostat) and ~20% of US businesses (Census). Large firms adopt far more than small (37% vs under 20% in the US).
- What is the AI adoption rate for small businesses?
- Small firms lag large ones sharply: in the US, AI adoption was about 37% among large firms versus under 20% among small firms (US Census Bureau, 2026). Cost, skills and integration are the main barriers — which is why fixed-scope, senior delivery matters most for smaller companies.
- Is AI adoption growing in 2026?
- Yes, quickly. EU enterprise adoption rose from 13.5% (2024) to 20% (2025), +6.5 points, per Eurostat; organizational adoption rose from 78% to 88% in executive surveys; and generative-AI adoption in marketing and sales roughly doubled (McKinsey). The frontier is shifting to AI agents — Gartner expects 33% of enterprise apps to include agentic AI by 2028.
- What share of AI value is actually captured?
- Despite high adoption, value capture lags: only ~23% of enterprises were scaling AI agents in 2025 (McKinsey), and Gartner expects 40%+ of agentic AI projects to be cancelled by 2027. The gap between adopting AI and shipping value in production is the central challenge.
- How many jobs will AI create or displace?
- The World Economic Forum forecasts 170 million new jobs and 92 million displaced by 2030 — a net gain of about 78 million — while 39% of core skills change. The IMF estimates ~40% of global jobs are exposed to AI (about 60% in advanced economies). Exposure means change, not necessarily loss.
- What is the ROI of enterprise AI?
- McKinsey finds most companies reporting a 10–20% cost or revenue impact where AI is applied by function, and estimates $2.6–4.4 trillion in annual value potential from generative AI. But value concentrates in a minority that reaches production; adoption alone does not deliver ROI.
- What are AI adoption rates by industry in 2026?
- Adoption varies widely by sector. Technology/IT and financial services lead; UK financial firms are at 75% and insurance at 95% (Bank of England/FCA), 90% of Fortune 100 use GitHub Copilot (Microsoft/GitHub), retail decision-makers are 55% ready to deploy gen AI (Google Cloud/NRF), while about 51% of manufacturers report using AI (NAM). See our dedicated AI adoption by industry report for the full breakdown.
- How many companies are actually succeeding with AI?
- Despite 88% adoption, only about 6% of organizations are AI 'high performers' capturing significant value, and roughly a third have scaled beyond pilots (McKinsey 2025). Yet 92% plan to increase AI budgets over the next three years — spending is rising faster than value capture.
- Is there a global AI index or ranking by country?
- Several exist. For enterprise adoption, Eurostat ranks EU countries (Denmark, Finland and Sweden lead); Stanford HAI's AI Index ranks countries on research, investment and models; and Affärslivet publishes a National AI Power Index. On pure enterprise adoption, the Nordics top Europe and the US leads on investment and models.
- How fast is AI adoption accelerating?
- Very fast: EU enterprise adoption jumped 6.5 points in a single year (13.5%→20%), and 26 of 27 EU countries recorded increases in 2025 (Eurostat). Organizational adoption rose from 78% to 88% year on year, and 92% of firms plan to increase AI budgets (McKinsey).
Going deeper on AI? We go deep on the leading AI consulting firms in Sweden, AI consultants across Europe and AI consultants in the Nordics — and we also compare the AI companies based in Stockholm.
More AI guides from Affärslivet
Consulting & choosing a firm
AI agents & frameworks
Generative AI, RAG & chatbots
Implementation & delivery
Governance & EU AI Act
Corporate AI training
By industry
Svenska djupdykningar
How we make these guides
Affärslivet's AI desk compiles these guides editorially: every market figure carries its own named primary source with a link (see each chart/table), vendors are assessed against publicly known focus, and cases are attributed to each company. Pages are kept current with a visible date.
Read next
AI Statistics 2026 — 230+ facts Best AI consulting firms AI Intelligence