Paloren compiles AI adoption statistics from cited public sources including McKinsey and Stanford HAI.
Paloren compiles AI adoption statistics from cited public sources including McKinsey and Stanford HAI. See what the figures measure and how to apply them without overstating the evidence.
For teams assessing AI adoption evidence.
AI adoption statistics show growing enterprise use. McKinsey reports nearly nine in ten survey respondents using AI in at least one function and 44 percent reporting enterprise-scale adoption. Stanford AI Index and OECD.AI provide complementary evidence.
Nearly nine in ten respondents report regular use in at least one function.
Inside the detail
McKinsey State of AI reports that nearly nine in ten respondents report regular use of AI in at least one business function. The same 2026 survey reports that 44 percent of respondents now say AI is scaling across their enterprise, up from 38 percent a year earlier.
It also reports that 40 percent of respondents from large organisations with annual revenues above USD 1 billion report scaling AI agents, up from 27 percent.
Keep these in view
Regular use in at least one function
44 percent enterprise scaling
40 percent large-organisation agents
They describe survey respondents, not all companies.
Inside the detail
These figures describe survey respondents and should not be applied to every business. Large organisations may adopt different systems from smaller companies. Regular use also differs from enterprise scaling.
Paloren recommends checking the survey population and definition before using any figure in an internal business case.
Keep these in view
Survey respondents
Different organisation sizes
Use versus scaling
| Finding | Population | Source |
|---|---|---|
| Nearly nine in ten use AI in at least one function | Survey respondents | McKinsey State of AI |
| 44 percent report enterprise-scale adoption | Survey respondents | McKinsey State of AI |
| 40 percent report scaling AI agents | Respondents from organisations with revenue above USD 1 billion | McKinsey State of AI |
| AI research, economy and policy data | Cross-source public evidence | Stanford AI Index |
Each figure describes its own survey population.
Source: Cited source pages above.
Stanford AI Index and OECD.AI.
Inside the detail
Stanford HAI publishes the AI Index, a public report tracking AI research, technology, economy and policy data. OECD.AI tracks public AI policies and initiatives. These sources do not answer the same question as McKinsey, but they provide complementary evidence on research, policy and system-level change.
Keep these in view
Stanford AI Index
OECD.AI policy navigator
Complementary evidence
Paloren turns cited adoption figures from McKinsey and Stanford HAI into usable evidence. The workflow below shows how a team can check what each statistic measures before quoting it.
| Stage | What changes |
|---|---|
| Signal | Record the McKinsey State of AI headline: nearly nine in ten respondents reporting regular AI use in at least one business function. |
| Synthesis | Separate regular use from enterprise scaling, noting that 44 percent of respondents report scaling across the enterprise, up from 38 percent a year earlier. |
| System | Check the survey population and definitions behind each figure, including the 40 percent of large organisations above USD 1 billion revenue reporting scaled AI agents. |
| Scale | Assign an analyst to track which figures are quoted in the internal business case and compare against OECD.AI policy data over time. |
A strategy team drafts a business case claiming most companies use AI. Signal captures the McKinsey figure of nearly nine in ten respondents using AI in one function. Synthesis notes that regular use in one function is not enterprise scaling, and that 44 percent scaling is a respondent share, not a market share.
System checks that the survey covers respondents, not every business, and that the 40 percent agent figure applies only to organisations above USD 1 billion revenue. Scale keeps the corrected wording in the case and rechecks figures against the next survey wave.
Written teaching example with hypothetical inputs, not a client result.
Operating metric: Share of quoted statistics whose survey population and definition were verified.
What you take forward
Assess our AI readiness
Tell Paloren the workflow and evidence available.
Adoption, scaling or agents.
Use original reports.
Enterprise, SME or public policy.
Avoid overgeneralising.
| Finding | Population | Source |
|---|---|---|
| Nearly nine in ten use AI in at least one function | Survey respondents | McKinsey State of AI |
| 44 percent report enterprise-scale adoption | Survey respondents | McKinsey State of AI |
| 40 percent report scaling AI agents | Respondents from organisations with revenue above USD 1 billion | McKinsey State of AI |
| AI research, economy and policy data | Cross-source public evidence | Stanford AI Index |
Each figure describes its own survey population.
Cited source pages above.
Adoption table
Source notes
Segment comparison
Business-case template
Review plan
Owner
| Step | Stage | What it changes |
|---|---|---|
| 01 | Signal | Record the McKinsey State of AI headline: nearly nine in ten respondents reporting regular AI use in at least one business function. |
| 02 | Synthesis | Separate regular use from enterprise scaling, noting that 44 percent of respondents report scaling across the enterprise, up from 38 percent a year earlier. |
| 03 | System | Check the survey population and definitions behind each figure, including the 40 percent of large organisations above USD 1 billion revenue reporting scaled AI agents. |
| 04 | Scale | Assign an analyst to track which figures are quoted in the internal business case and compare against OECD.AI policy data over time. |
The figure depends on the source and population. McKinsey reports nearly nine in ten respondents using AI in at least one function.
McKinsey reports 44 percent of respondents say AI is scaling across their enterprise in its 2026 survey.
McKinsey reports 40 percent among respondents from organisations with annual revenues above USD 1 billion.
No. Adoption varies by organisation size, sector, workflow and definition.
What does adoption mean in your team?
Assess our AI readiness
