Global AI Adoption Trends by Industry Sector, Country, and Business Function (2017–2025): A Compiled Research Dataset
Description
This dataset compiles and synthesises publicly available artificial intelligence (AI) adoption and growth indicators from leading institutional research reports, spanning the years 2017 to 2025. It is structured across seven thematic dimensions, covering organisational AI adoption rates, business function-level deployment, global AI tool user milestones, private AI investment by country, industry-sector adoption rates, public sentiment toward AI, and key headline KPIs. Data is sourced from and attributed to: McKinsey & Company Global Survey on AI (2022–2025), Stanford HAI Artificial Intelligence Index Report 2025, OpenAI official announcements, GitHub/Microsoft earnings disclosures, Ipsos Global AI Sentiment Survey 2024, World Bank South Asia AI Report 2025, IBM AI Adoption Index 2024, Oxford Insights Government AI Readiness Index 2024, and SimilarWeb platform analytics. The dataset is intended to support researchers, data analysts, and policymakers working on AI trend analysis, digital transformation studies, technology policy, and sector-level AI readiness assessments. All figures are either directly verified from primary sources or clearly labelled as modelled estimates anchored to verified data points. Source attribution is embedded within the dataset at the row level. Files are provided in both .xlsx (multi-sheet, formatted workbook) and .csv formats for compatibility with tools such as Microsoft Power BI, Tableau, R, and Python.
Files
Steps to reproduce
This dataset was compiled through systematic review and extraction of publicly available AI research reports and institutional surveys. The following steps describe the data collection and compilation methodology: Step 1 — Source Identification Primary sources were identified based on institutional credibility and data availability. Sources include: McKinsey & Company Global Survey on AI (2022–2025), Stanford HAI Artificial Intelligence Index Report 2025, OpenAI official public announcements, GitHub/Microsoft earnings disclosures, Ipsos Global AI Sentiment Survey 2024, World Bank South Asia AI Report 2025, IBM AI Adoption Index 2024, Oxford Insights Government AI Readiness Index 2024, ITU/GSMA internet user statistics, and SimilarWeb platform analytics. Step 2 — Data Extraction Quantitative figures (adoption percentages, user counts, investment values, sentiment scores) were extracted directly from the above reports. Each figure was tagged with its originating source, report year, and applicable geographic or sector scope. Step 3 — Verification & Classification Each data point was classified into one of three reliability tiers: (a) Verified — directly quoted from a primary source publication; (b) Modelled/Anchored — interpolated or extrapolated using verified anchor points and documented scaling methodology; (c) Projected — forward estimates based on confirmed growth trajectories. This classification is embedded in the dataset's "Data_Type" column. Step 4 — Structuring Data was organised into seven thematic tables covering: organisational AI adoption over time, business function-level deployment, AI tool user milestones, private investment by country, industry sector adoption rates, public sentiment by country, and headline KPIs. Each table includes inline source citations at the row level. Step 5 — File Preparation Final dataset was exported in .xlsx format (Microsoft Excel, multi-sheet workbook) and .csv format (one file per thematic table) for compatibility with Power BI, Tableau, R, Python (pandas), and Google Sheets. No proprietary software or statistical package is required to open or use these files. Reproducibility Note: As all primary sources are publicly accessible institutions, any researcher can independently verify individual figures by consulting the cited reports. Direct URLs to primary sources are embedded within the dataset.