Home/Adoption Trends/Enterprise AI Adoption Year-by-Year
Adoption Trends

Enterprise AI Adoption Year-by-Year

McKinsey's State of AI surveys, conducted annually since 2017, form the longest-running longitudinal record of enterprise AI adoption. The 2025 edition surveyed 1,993 participants across 105 countries and found 88% of organizations now use AI in at least one business function — up from 20% in the first survey eight years ago. But the same dataset shows that only 6% of organizations qualify as "AI high performers" with measurable enterprise impact. This page tracks both the adoption and the impact-gap year by year.

4 4 visualizations + signals card 7 sources Last updated June 2026 Free to embed
Loading chart...
Loading chart...
Loading chart...
Chart 1 · The 8-year arc

Enterprise AI adoption by year, 2017–2025

Loading chart...
Source: McKinsey "The State of AI" annual surveys, 2017–2025. Sample sizes ranged from approximately 1,500 to 2,000 participants per year across 100+ countries. The metric is the share of organizations reporting they use AI in at least one business function. Slight methodological adjustments between 2020–2022 explain minor year-to-year fluctuations.
Chart 2 · The gen AI subset

Generative AI use specifically, 2022–2025

Loading chart...
Source: McKinsey State of AI 2023, 2024, 2025 reports. Generative AI use specifically (distinct from all AI). The 2023 baseline (33%) was the first year McKinsey separated gen AI from broader AI in the survey.
Chart 3 · Adoption vs scaling

The scaling gap, 2025

Loading chart...
Source: McKinsey State of AI 2025. 88% adoption refers to any AI use in at least one function. 33% enterprise-scaling = organizations scaling AI across the enterprise. 39% any-EBIT-impact = organizations reporting any EBIT impact from AI. 6% high performers = >5% of EBIT attributed to AI.
Chart 4 · The agentic frontier

AI agent deployment status, mid-2025

Loading chart...
Source: McKinsey State of AI 2025 (survey fielded June–July 2025). 62% of organizations report experimenting with AI agents; 23% report scaling agentic AI in at least one function; "Most scaling organizations are doing so in only one or two functions," per McKinsey.

About this data

The AI Behavior Index is the research arm of OneChat AI. This page is built almost entirely on a single dataset — McKinsey's annual State of AI surveys — because it is the longest-running, most consistent enterprise-AI tracking dataset publicly available. McKinsey has surveyed approximately 1,500–2,000 enterprise participants annually since 2017, across 100+ countries and all major industries.

One important caveat: McKinsey's methodology has been refined several times over the eight-year period, particularly between 2020 and 2022. The headline "AI adoption" metric is the share of organizations using AI in at least one business function — a relatively low bar. Deeper metrics (scaling, EBIT impact, high performer status) are stricter and tell a more nuanced story, which is why this page includes both the headline curve and the depth metrics. Data is refreshed annually when McKinsey publishes the new edition. If you have a study to suggest or notice an error, contact us at research@aibehaviorindex.org.

For Journalists & Researchers

Use this data in your work.

Every statistic, chart, and graphic in this index is free to use and cite, with full source attribution. Can’t easily find what you need? Use our search bar to search by keyword, topic, or category.

✉️
Talk to our research team
Need a specific cut of data, an interview, or a quote? Email us — we typically respond within one business day.
research@aibehaviorindex.org →

How the data works

Every statistic shown is sourced from a publicly available study, survey, or report. We aggregate, organize, and contextualize this data — but the underlying research is conducted by the cited sources. Click any source link to access the original methodology. If you run into any issues or have a study to suggest, contact us at research@aibehaviorindex.org.