Artificial Intelligence Statistics, Trends & Data (Updated for 2026)

Artificial intelligence has moved well beyond experimentation. In 2026, it influences how businesses operate, how software gets built, and how work gets done across nearly every industry. Within the broader AI ecosystem, generative AI, agentic AI solution, computer vision, and machine learning are all advancing - with generative AI still leading public and enterprise adoption, and agentic AI now the fastest-growing category behind it.

Organizations are increasingly measuring AI through business outcomes rather than experimentation: faster workflows, lower operating costs, better customer experiences, and measurable productivity gains. At the same time, investment in AI infrastructure, automation, and intelligent agents continues to accelerate as businesses look for durable competitive advantages.

The statistics below bring together the latest verified data on AI adoption, investment, market growth, enterprise usage, workforce impact, and generative and agentic AI trends. Every figure is attributed to a named, dated primary source - official statistical agencies where available, named analyst or survey research everywhere else.

AI Statistics at a Glance

Global adoption figures vary widely depending on how "AI use" is defined and measured. The numbers below show both the official statistical view and the broader self-report view side by side - a distinction that matters more than most single-number headlines let on.

• 19.95% of EU enterprises (10+ employees) used at least one AI technology in 2025, up from 13.48% in 2024 (Eurostat, 2025).
• 20.2% of OECD firms reported using AI in 2025, more than doubling from 8.7% in 2023 (OECD, January 2026).
• 88% of organizations now regularly use AI in at least one business function - up from 78% a year earlier (Stanford HAI, AI Index Report 2026, published April 2026).
• 53% global population adoption of generative AI within three years of ChatGPT's release - faster than the PC or the internet reached the same milestone (Stanford HAI, AI Index Report 2026).
• 24% of organizations report full-scale, enterprise-wide AI adoption in 2026, up from 12% in 2025 (TEKsystems, 2026).

The gap between the official ~20% figure and the self-report ~88% figure isn't a contradiction - it reflects two different questions. Eurostat and OECD measure integrated, production-level use across a representative business sample. Stanford HAI and most enterprise surveys measure "any use in any function," which captures pilots, individual tool subscriptions, and shadow AI use. Both are accurate; they're just not the same yardstick.

AI Market & Investment Statistics

Market-size figures differ sharply depending on scope - hardware vs. software vs. total spend, narrow API revenue vs. infrastructure-inclusive definitions. The numbers below are kept narrow and attributed precisely rather than blended into one headline figure, because that blending is where most "AI market size" claims online quietly stop being verifiable.

Global AI Market Size

• Worldwide AI spending is forecast to reach $632 billion by 2028, a 29.0% five-year CAGR from 2024 (IDC Worldwide AI and Generative AI Spending Guide, 2024).
• Worldwide generative AI spending specifically is forecast to reach $644 billion in 2025, a 76.4% increase over 2024, with hardware accounting for 80% of that spend (Gartner, March 2025).
• US private AI investment reached $109.1 billion in 2024 - nearly 12 times China's $9.3 billion (Stanford HAI AI Index Report 2025).

A note on generative-AI-specific market sizing: published 2026 estimates for this segment alone range from $28.45 billion (Mordor Intelligence) to $394.66 billion (Statista) - a roughly 14x spread for the same calendar year, driven entirely by scope definition rather than genuine disagreement.

Narrow "software and API" definitions land near $30–80 billion; infrastructure-inclusive definitions land near $160–400 billion. 

Enterprise generative AI spending reached $37 billion in 2025, up from $11.5 billion in 2024 - a 3.2x year-over-year increase, split roughly $19B in applications and $18B in infrastructure, with foundation model APIs alone accounting for $12.5B (Menlo Ventures, State of Generative AI in the Enterprise, December 2025).

AI Startup Funding & Unicorns

• AI startups raised $202.3 billion in 2025 - close to 50% of all global venture capital, up more than 75% year-over-year from $114 billion in 2024 (Crunchbase News, published December 15, 2025).
• Momentum accelerated sharply into 2026: AI startups raised $242 billion in Q1 2026 alone, roughly 80% of all global venture funding that quarter (Crunchbase News, April 1, 2026).
• H1 2026 set an all-time record: global venture funding hit $510 billion, already surpassing the full $440 billion invested in all of 2025 (Crunchbase News, July 2026).
• AI is now the dominant force in unicorn creation: 215 AI unicorns globally, up 87 in a single year and accounting for 36% of total unicorn value - the highest share of any sector (Hurun Research Institute, Global Unicorn Index 2026, published June 25, 2026).

Corporate & Infrastructure Spending

• Average enterprise AI spending reached $207 million per organization among large companies surveyed in Q1 2026 (KPMG AI Quarterly Pulse Survey, 2026).
• About 78% of enterprises increased their AI budget year over year (PwC AI Business Predictions, 2026); separately, 68% of CIOs planned to raise AI budgets in 2026 (Morgan Stanley CIO Survey, Q1 2025).
• The AI data center GPU market is valued at $45.04 billion in 2026 (Mordor Intelligence, AI Data Center GPU Market report, 2026).
• Capital expenditure by five large technology companies exceeded $400 billion in 2025 and is projected to rise a further 75% in 2026 - now larger than global investment in oil and natural gas production (International Energy Agency, Key Questions on Energy and AI, April 2026).

Government & Regulatory Investment Signal

The clearest government-side signal available is regulatory rather than financial: the EU AI Act's high-risk system obligations apply from August 2, 2026 (Regulation (EU) 2024/1689; EUR-Lex, 2024). A defensible global government AI spending total isn't publicly available at benchmark quality as of this writing - treat any single number you see elsewhere with caution.

Enterprise AI Adoption Statistics

Adoption Rates

• 19.95% of EU enterprises and 20.2% of OECD firms used AI in 2025 on the official measure (Eurostat, 2025; OECD, January 2026).
• 88% self-report using AI in at least one business function - the broader "any use" measure (Stanford HAI AI Index Report 2026, April 2026).
• 24% of organizations report full-scale enterprise-wide adoption, with digital leaders at 38% versus 9% for laggards (TEKsystems, 2026).
• 18% of US firms had adopted AI by year-end 2025 on the Federal Reserve's production-use measure, while 39% of US workers already used AI at work - a gap driven largely by unsanctioned "shadow AI" use (Federal Reserve Board; Federal Reserve Bank of New York, 2026).

Read more: 5 Enterprise AI Trends Reshaping Business

Adoption by Company Size

• 55.03% of large EU enterprises (250+ employees) use AI, versus 30.36% of medium and 17.00% of small enterprises (Eurostat, 2025).
• The OECD shows a near-identical pattern: 52.0% of large firms versus 17.4% of small firms (OECD, 2025).
• US small-business AI use rose from about 4.2% to 5.7% between September 2023 and September 2024, led by chatbots and digital assistants (U.S. Census Bureau, December 2024).

Adoption by Industry

(Eurostat, 2025; OECD, 2025)

Within EU adopters specifically, AI is applied most often to marketing/sales (34.70%) and business administration (31.05%), followed by production processes (26.40%) and logistics (19.30%) (Eurostat, 2025).

Adoption by Country

• EU: 19.95% (Eurostat, 2025); OECD average: 20.2% (OECD, 2025) - both up sharply from 2023 baselines.
• Canada: 12.2% of businesses used AI in the prior 12 months as of Q2 2025, exactly double the 6.1% rate a year earlier (Statistics Canada, 2025).
• United States: 5.4% on the Census Bureau's narrow "previous two weeks" measure (February 2024), versus 18% on the Federal Reserve's broader year-end 2025 measure - the two aren't directly comparable, and neither should be quoted without its measurement window.
• United Kingdom: 9% of firms (2023 data). Among EU member states, Denmark (27.6%), Sweden (25.0%), and Finland (24.7%) lead (Eurostat, 2025).

Adoption Barriers

• 70.89% of EU enterprises that considered AI but didn't adopt it cite lack of relevant expertise as the top barrier; 52.52% cite legal or regulatory uncertainty (Eurostat, 2025).
• In the UK, 39% of firms cite difficulty identifying use cases as their top barrier, ahead of cost at 21% (ONS, 2023 data).
• In Canada, 78.1% of businesses with no adoption plans cite AI as "not relevant" to their operations - a materially different barrier pattern than the EU or UK (Statistics Canada, 2025).
• Only 15.9% of US workers report employer-provided AI training, despite 39% already using AI at work - a 23-point training gap (Federal Reserve Bank of New York, April 2026).

This expertise gap is precisely where most AI initiatives stall before they ever reach a pilot - not because the technology isn't ready, but because no one on staff has mapped which workflows are worth automating first. That's the gap Ciphernutz's AI Readiness Audit is built to close before a company commits budget to a build.

Generative AI Statistics

Adoption

• 71% of organizations regularly use generative AI in at least one function, up from 33% in early 2023 (McKinsey State of AI, 2025).
• On the stricter official measure, only about 6.5% of EU enterprises use AI specifically for generating text, images, or other content (Eurostat, 2025), and 8.4% of Canadian AI-using businesses do the same (Statistics Canada, Q2 2025).
• 67% of enterprises were piloting generative AI and 18% had scaled it, per a Q4 2024 C-suite survey of over 2,700 executives (Deloitte, State of GenAI in the Enterprise).

Enterprise Usage by Function

McKinsey's self-report data and the EU's stricter official measure tell a consistent story about where generative AI lands first, even though the absolute levels differ:

Marketing and sales leads on both measures - the single most consistent finding in this entire dataset.

Investment

• Global private investment in generative AI reached $33.9 billion in 2024, up 18.7% from 2023 (Stanford HAI AI Index Report 2025).
• 215 newly funded generative AI companies were founded in 2024 (Stanford HAI AI Index Report 2025).
• Enterprise generative AI revenue grew from $1.7 billion in 2023 to $37 billion in 2025 - the fastest-scaling software category on record, now roughly 6% of the global SaaS market (Menlo Ventures, December 2025).

Consumer Adoption

• ChatGPT remains by far the largest consumer AI product - 2.7x larger than #2 Gemini on web traffic and 2.5x larger on mobile monthly active users. Weekly active users grew by 500 million over the prior year to approximately 900 million (Andreessen Horowitz, Top 100 Gen AI Consumer Apps, 6th Edition, published March 2026, data as of January 2026).
• Claude and Gemini are the fastest-growing challengers: Claude's US paid subscribers grew over 200% year-over-year and Gemini's grew 258%, as of January 2026 (Yipit Data via a16z, March 2026).
• Across OECD member countries, individual AI use rose from 19% (2023) to 28% (2024) to roughly 37% (2025) (OECD, January 2026).

User-level and enterprise-level adoption figures measure different things and shouldn't be combined into a single rate.

Agentic AI Statistics

Agentic AI is the fastest-moving category in 2026, and most usable data still comes from vendor and analyst surveys rather than official statistics - no statistical agency yet tracks "AI agents" as a distinct category the way Eurostat tracks generative AI use.

Market Growth

The best available real-usage evidence comes from a16z's analysis of more than 100 trillion tokens of production LLM traffic through OpenRouter (a16z, State of AI, published December 2025):

• Agentic inference - workflows where a model plans, calls tools, and iterates across multiple steps rather than answering a single prompt - is the fastest-growing usage pattern on the platform.
• Prompt lengths and session turn counts are both rising, and reasoning- and tool-use-specialized models are gaining share faster than general chat models.
• OpenRouter alone grew from roughly 10 trillion tokens processed per year to over 100 trillion within about 18 months, now serving 5+ million developers across 300+ models from 60+ providers.

Separately, Gartner forecasts that 33% of enterprise software applications will include agentic AI by 2028, up from less than 1% in 2024, and that agentic AI will handle 15% of day-to-day work decisions autonomously by the same date (Gartner, 2024).

Enterprise Agent Adoption

• 63% of IT leaders report piloting or deploying AI agents, per a survey of 1,800 IT leaders (Salesforce State of IT, August 2024).
• 79% of senior executives report AI agents being adopted somewhere in their company, per a survey of 300 senior executives (PwC AI Agent Survey, May 2025).
• 10% of organizations already use AI agents in production, and 82% plan to integrate them within one to three years, per a survey of 1,100 executives (Capgemini Research Institute, 2024).
• Deloitte projects 25% of enterprises already using generative AI will deploy AI agents in 2025, rising to 50% by 2027 (Deloitte TMT Predictions, 2025).
• Despite that adoption curve, only about 1% of leaders describe their generative AI rollouts as reaching agentic "maturity" (McKinsey State of AI, 2025) - adoption and maturity are still two very different milestones.

Agent Investment

74% of executives report positive ROI from AI agent investments within 12 months - the strongest available proxy for agent-specific returns (Google Cloud, ROI of AI Agents, 2025). A clean, benchmark-grade total for agent-specific venture investment (distinct from broader AI funding) isn't publicly available yet; treat any single dollar figure you see for "agent funding" as a directional estimate rather than a verified total.

Business Impact & ROI Statistics

The clearest, most causally robust AI evidence available in 2026 sits at the worker-task level - controlled field experiments with real before/after measurement. Firm-level gains are also documented; economy-wide gains remain harder to attribute and are still debated among economists.

Productivity

• AI assistant access increased customer-support throughput by 14% on average, and by 34% for novice and low-skilled workers specifically - with statistically zero gain for top performers - across 5,179 agents at a Fortune 500 firm (Brynjolfsson, Li & Raymond, NBER working paper 31161, published in the Quarterly Journal of Economics, 2023/2025).
• AI coding assistants raised completed developer tasks by 26.08% across 4,867 developers at three companies, including Microsoft and Accenture (Cui, Demirer, Jaffe, Musolff, Peng & Salz, NBER working paper 33777, 2025).
• European AI-adopting firms show a 4% short-run labor productivity gain with no adverse short-run employment effect, across more than 8,800 firms in 25 countries (BIS/EIB working paper, January 2026).
• Within the AI "jagged frontier," consulting tasks were completed 25% faster with 40% higher human-rated quality (Dell'Acqua et al., Harvard Business School/BCG, 758 consultants, 2023) - but outside that frontier, gains disappear or reverse: in a Kenya field experiment, high-performing entrepreneurs gained 20% while lower performers lost 10%, netting a zero average effect (HBS/Berkeley, 2023).

Revenue & Cost Impact

• 21% of organizations reported material generative AI impact on earnings (over 5% of EBIT) in 2026, up from 11% in the prior wave (McKinsey State of AI, 2026).
• About 50% of organizations in AI-adopting functions report a revenue uplift, and 33% report a cost reduction (McKinsey State of AI, 2025).
• Top-quartile AI use cases return roughly 3.5x on investment (McKinsey State of AI, 2024).
• Only about 25% of companies report having captured significant value from AI investments overall - the widely cited "AI impact gap" (BCG AI Radar, 2025).

Deployment Reality

• Nearly 95% of enterprise generative AI pilots fail to reach measurable production impact, with data quality and workflow integration cited as the leading blockers (MIT NANDA, State of AI in Business, 2025).
• At least 30% of generative AI projects were abandoned at the proof-of-concept stage by the end of 2025 (Gartner, forecast published 2024).
• Only 21% of organizations report material EBIT impact from generative AI, underscoring that adoption and realized value are two separate milestones (McKinsey State of AI, 2026).

AI Workforce Statistics

AI is reshaping hiring at a pace the labor-market data providers themselves describe as unmatched by any prior technology cycle - but the effect is concentrated in specific roles and skills rather than spread evenly across the workforce.

• AI Engineer is the #1 fastest-growing job title in the United States for the second consecutive year, with postings up 143% year-over-year in 2025 (LinkedIn, Jobs on the Rise 2026).
• The global economy added 1.3 million new AI-related jobs over the past two years - including AI Engineers, Forward-Deployed Engineers, and Data Annotators, plus over 600,000 new AI-enabled data center roles (LinkedIn Economic Graph data, via World Economic Forum, published January 2026).
• Python remains the single most in-demand AI-adjacent skill, appearing in 258,674 postings - up 391% from the 2013–15 baseline and nearly 30% year-over-year. Lightcast added "Agentic AI" as a new tracked skill cluster this year, bringing the total to 10 clusters and 300+ AI skills monitored (Lightcast, contributing to Stanford HAI AI Index Report 2026, April 2026).
• Workers with in-demand AI skills earned a 56% average wage premium in 2026, more than double the 25% premium recorded a year earlier (PwC Global AI Jobs Barometer, 2026).
• Only about 43% of US workers report regularly using AI at work, even as AI-related job postings grow against an otherwise flat hiring market (Indeed Hiring Lab, January 2026).
• Only 15.9% of US workers report employer-provided AI training, despite 39% already using AI on the job - a 23-percentage-point gap between usage and formal readiness (Federal Reserve Bank of New York, April 2026).

Industry-Specific AI Statistics

Healthcare

• 71% of nonfederal acute care hospitals used predictive AI integrated into their electronic health records in 2024, up from 66% in 2023 (ONC/ASTP Data Brief, U.S. Department of Health and Human Services, September 2025).
• 66% of US physicians reported using health AI tools in 2024, up from 38% in 2023 - a 78% relative increase (American Medical Association, Augmented Intelligence Survey, 2024–25).

Retail

• 86% of US retailers already have AI governance policies in place, and 93% plan to develop or continue developing them within the next 12 months (National Retail Federation, Center for Digital Risk & Innovation, survey of 56 US retail AI leaders, summer 2025, published December 2025).
• 77% of retailers allocate 5% or less of their tech budget to AI today, but 39% expect AI to exceed 10% of tech spend within three years (NRF, December 2025).
• Globally, 45% of consumers use AI somewhere in their shopping journey - 41% for product research, 33% to interpret reviews, and 31% to hunt for deals (IBM Institute for Business Value with NRF, global survey of 18,000+ consumers across 23 countries, Q3 2025, published January 2026).

Finance

• Financial and insurance services show roughly 28% AI adoption among OECD firms and 26.9% among EU enterprises (OECD, 2025; Eurostat, 2025).
• Finance and high-skill services are expected to see labor productivity gains of 2 percentage points or more in 2026, up from an implied 0.8 points in 2025 (Atlanta Fed Policy Hub, March 2026).

Manufacturing

Manufacturing shows roughly 18% AI adoption among OECD firms and 17.5% among EU enterprises (OECD, 2025; Eurostat, 2025) - trailing knowledge-intensive sectors but accelerating fastest in maintenance, quality control, and engineering documentation use cases (OECD AI Outlook, 2025).

Marketing

Marketing and sales is the top functional use case for generative AI on both the self-report measure (65%, McKinsey, 2025) and the official EU measure (34.70% of AI-adopting enterprises, Eurostat, 2025) - the most consistent cross-methodology finding in this dataset.

Real Estate

14.5% of EU enterprises in the real estate sector use AI, per Eurostat's NACE industry breakdown (Eurostat, 2025) - placing the sector below EU knowledge-intensive industries but roughly in line with administrative and support services. A dedicated real-estate-specific ROI or use-case adoption statistic beyond this figure isn't available at benchmark quality yet.

Logistics

19.30% of EU AI-adopting enterprises use AI for logistics specifically - the fourth most common official use case after marketing/sales, business administration, and production processes (Eurostat, 2025).

LLM & AI Infrastructure Statistics

Foundation Model Usage

Among US enterprises, Anthropic holds 40% of enterprise LLM API spend, OpenAI holds 27%, and Google holds 21% - the top three providers together account for 88% of enterprise LLM API usage (Menlo Ventures enterprise survey of ~500 US decision-makers, published December 2025). This is a reversal from 2023, when OpenAI held 50% of enterprise spend, Anthropic 12%, and Google 7%.

Token Usage & Compute

• OpenRouter's production traffic analysis of over 100 trillion tokens found reasoning- and tool-use-specialized models steadily gaining share over general chat models, alongside consistently rising prompt lengths and session turn counts (a16z, State of AI, December 2025).
• Open-source models - particularly reasoning-forward releases like DeepSeek R1 and Kimi K2 - are capturing meaningful share on cost efficiency and flexibility grounds (a16z, December 2025).

Energy & Data Center Growth

• Global data centre electricity demand grew 17% in 2025, far outpacing the 3% growth in overall global electricity demand. AI-focused data centres grew even faster - electricity consumption up 50% in a single year (International Energy Agency, Key Questions on Energy and AI, April 2026).
• Global data centre electricity consumption is projected to roughly double by 2030, reaching approximately 945–950 TWh; AI-focused data centre consumption specifically is set to triple over the same period (IEA, April 2026).
• Energy use per individual AI task has dropped by at least an order of magnitude annually - but aggregate demand keeps climbing because user growth and heavier tasks (video generation, agentic reasoning) are outpacing those efficiency gains (IEA, April 2026).

AI Governance, Risk & Security Statistics

Regulation

• The EU AI Act (Regulation (EU) 2024/1689) was published July 12, 2024, with general-purpose AI obligations applying from August 2, 2025, and high-risk system obligations from August 2, 2026 (EUR-Lex, 2024).
• The Colorado AI Act - the first US state law imposing risk-management duties on AI developers and deployers - took effect in February 2026.
• ISO/IEC 42001, the first certifiable AI management system standard, was published December 2023 (ISO, 2023). The NIST AI Risk Management Framework 1.0 followed in January 2023, with a companion Generative AI Profile added in July 2024 (NIST, 2023/2024).

Risk

• 70.89% of EU enterprises that considered AI but didn't adopt it cite lack of expertise as a barrier, and 52.52% cite legal uncertainty (Eurostat, 2025) - see Adoption Barriers above for the full breakdown.
• IMF analysis suggests binding EU AI Act regulatory constraints could reduce projected AI productivity gains by roughly 30% (IMF, 2025).
• Publicly reported AI-related security incidents increased 56.4% from 2023 to 2024 (Stanford HAI AI Index Report 2025).

Security

The OWASP Top 10 for LLM Applications (2025 edition) names prompt injection, sensitive information disclosure, and supply-chain vulnerabilities as the leading risks in enterprise generative AI security baselines (OWASP, 2025).

A rigorously sourced, benchmark-grade figure for AI-specific breach frequency or cost - distinct from general data-breach statistics - isn't publicly available yet; most numbers circulating for this specific metric trace back to marketing content rather than disclosed-methodology surveys, so we're leaving it out rather than citing something that won't hold up.

Future AI Trends (2027 and Beyond)

• Gartner forecasts 33% of enterprise applications will include agentic AI by 2028, up from less than 1% in 2024 (Gartner, 2024).
• Deloitte projects 50% of generative-AI-using enterprises will deploy AI agents by 2027, up from 25% in 2025 (Deloitte TMT Predictions, 2025).
• Macro productivity gain estimates for the next decade range from +0.07 to +1.3 percentage points per year, depending on task-coverage and adoption-pace assumptions (Acemoglu, NBER, 2024; Aghion & Bunel, OECD, 2025).
• Worldwide AI spending is forecast to reach $632 billion by 2028 (IDC, 2024), while capex from the five largest tech companies alone is set to rise a further 75% in 2026 on top of 2025's $400 billion (IEA, April 2026).

Where This Leaves Most Businesses

Taken together, these figures point to a market where adoption is real but maturity is rare - 88% of organizations use AI somewhere, but only 1% call their generative AI rollout mature, and nearly 95% of pilots never reach production. The gap isn't the technology. It's the absence of a structured path from pilot to workflow to measurable outcome.

That's the gap Ciphernutz works in - from the initial AI Readiness Audit through AI Workflow Automation that are scoped to actually reach production. If you're trying to figure out where your organization sits against these numbers, get in touch - we'll tell you honestly whether you're ready to build or still need to close the readiness gap first.