Updated November 2026
AI has moved from experiment to everyday tool faster than almost any technology in business history. But the numbers behind that story are scattered across dozens of reports, and they don’t always agree. We pulled together the most useful AI statistics for business from the research leaders actually cite: McKinsey, Deloitte, Accenture, PwC, BCG, NVIDIA, Microsoft, Gartner, Stanford, the World Economic Forum, Gallup, KPMG, the U.S. Census Bureau, plus reporting from Forbes, Fortune and The Wall Street Journal.
Every figure below links to its source. Where studies measure different things (large enterprises versus all U.S. businesses, for example), we say so, because that is usually why two “AI adoption” numbers look so different.
- Nearly 9 in 10 organizations now use AI regularly, but only about 6% are getting a significant share of their earnings from it (McKinsey).
- Companies put roughly 93% of their AI budgets into technology and 7% into people, culture and training (Deloitte via Fortune).
- Insufficient worker skills is the top barrier to integrating AI, according to Deloitte’s survey of more than 3,000 leaders (Deloitte).
- Access to workplace learning fell from 59% to 51% of workers in a year, even as AI use rose (PwC).
- Jobs that require AI skills pay a 62% wage premium on average (PwC).
AI Adoption Statistics
Among larger organizations, AI is close to universal. Across all businesses, it is still the minority. Both statements are true, and the gap between them is where most of the opportunity sits.
organizations report regular AI use in at least one business function.
McKinsey, State of AI 2026of organizations used AI in 2025, up from 55% in 2023.
Stanford AI Index 2026of companies with 1,000+ employees are actively using AI, versus 64% of all respondents.
NVIDIA State of AI 2026of all U.S. businesses use AI to produce goods or services or run a business function.
U.S. Census Bureau, May 2026- Scaling is still the exception. 44% of McKinsey’s respondents say AI is scaling across their enterprise, up from 38% the year before (McKinsey).
- Size matters. 37% of U.S. firms with 250 or more employees use AI, compared with 32% of firms with 100 to 249 employees and far lower rates among the smallest businesses. Information (39.7%) and finance and insurance (33.9%) lead all sectors (U.S. Census Bureau).
- Adoption is fast by historical standards. Generative AI reached 53% population adoption within three years, faster than the personal computer or the internet (Stanford HAI).
- Employees are ahead of employers. 64% of workers used AI at work in the past year, and daily use of generative AI rose from 14% to 22% (PwC Global Workforce Hopes and Fears Survey 2026).
- Daily use is still rare in the U.S. workforce. 15% of U.S. employees use AI daily and 30% use it a few times a week or more. Only 25% say their organization has communicated a clear plan for AI (Gallup).
- Access is widening. The share of workers with access to company-sanctioned AI tools grew from under 40% to about 60% in a single year (Deloitte).
AI ROI and Business Impact Statistics
The headline finding of 2026 research is a value gap. Almost everyone uses AI. Far fewer can point to results on the income statement.
- Some impact, rarely large impact. 37% of organizations attribute at least some EBIT impact to AI, and about 6% qualify as AI high performers, both unchanged from the prior year (McKinsey).
- High performers redesign work. Nearly three-quarters of AI high performers have fundamentally redesigned workflows, compared with about a quarter of other organizations (McKinsey).
- Sustained value is limited. Only 25% of C-suite leaders report widespread and sustained value from AI, though 53% expect measurable ROI within 12 months (Accenture Pulse of Change, October 2026).
- Revenue ambitions outrun results. 74% of leaders hope to grow revenue through AI, but only 20% say they are already doing so (Deloitte).
- Where AI is deployed, leaders report gains. 88% of NVIDIA’s respondents say AI increased annual revenue in at least part of their business, and 87% say it reduced annual costs (NVIDIA).
- Most pilots stall. MIT’s NANDA initiative found about 95% of enterprise generative AI pilots delivered little or no measurable P&L impact, and traced the problem to a “learning gap” rather than model quality (Fortune).
- Strategy beats tools. A clear AI strategy raises measurable business impact by about 25 points, while better tools alone add about 5 (BCG AI at Work 2026).
- Work design doubles the odds. Organizations that prioritize work design are twice as likely to exceed their AI ROI expectations (Deloitte Global Human Capital Trends 2026).
AI Investment Statistics
- Record spending. Global corporate AI investment hit a record $581.69 billion in 2025 (Stanford AI Index 2026).
- More to come. 75% of C-suite leaders plan to increase AI investment over the next 12 months (Accenture), 86% of NVIDIA’s respondents expect their AI budgets to rise in 2026 (NVIDIA), and 60% of McKinsey’s respondents expect to increase AI investment next year (McKinsey).
- The people gap. Deloitte’s CTO has described a 93/7 split: 93% of AI budgets go to technology such as models, chips and software, and 7% to people, culture, workflow and training (Fortune).
- Training budgets are under pressure. More than half (53.7%) of HR executives in one recent survey said their organizations would accept, or had already accepted, slower employee skill development in exchange for greater AI investment (Forbes).
- Build versus buy is shifting. 32% of organizations decided against buying at least one software product because they could build it in-house with agentic coding tools (McKinsey).
AI Training and Skills Statistics
If there is one theme that runs through every major 2026 study, it is that skills, not software, now limit what companies get from AI.
of employees have received any AI training.
KPMG and University of Melbourneof workers have adequate access to learning and development, down from 59%.
PwC Hopes and Fears 2026of employers close AI skills gaps mainly by upskilling their own people.
Randstad via Forbesof employees say they can’t apply their AI training to daily work.
Docebo via Forbes- Skills are the top barrier. Insufficient worker skills is the leading obstacle to integrating AI into existing workflows, and educating the broader workforce to raise AI fluency is the most common talent response, cited by 53% of leaders (Deloitte).
- Training is uneven. 82% of enterprise leaders say they provide some AI training, but only 35% have a mature, organization-wide AI upskilling program (DataCamp). In a global survey of nearly 1,300 workers, 28.3% said their organization offers no AI training at all, and only about 1 in 3 had joined employer-provided AI training in the past six months (The Conference Board).
- Hiring over building. To close AI skills gaps, 63% of employers engage flexible talent and 33% hire permanent staff, while just 3% prioritize upskilling their existing people (Forbes).
- Training drives usage. Employees who receive at least five hours of AI training, plus in-person training and coaching, are far more likely to become regular users. Only about a third of employees said they had been properly trained (BCG AI at Work 2025).
- Training builds trust. Workers who received hands-on AI training or workshops reported 144% higher trust in their employer’s AI (Deloitte TrustID via Fortune).
- Many feel unprepared. Only 40% of employees say their training prepared them for how AI is changing their roles (Accenture Pulse of Change, January 2026 wave), and 48% say AI tools change so fast that it’s hard to keep current (Accenture, October 2026).
- Skills are changing quickly. 39% of workers’ existing skill sets will be transformed or outdated between 2025 and 2030, and 85% of employers plan to prioritize upskilling. If the world’s workforce were 100 people, 59 would need training by 2030 (World Economic Forum).
- Big employers are moving. Walmart is offering free AI training and certification to 1.6 million U.S. and Canadian employees through a partnership with Google (Fortune).

AI Productivity Statistics
- A full day back each week. 42% of regular frontline AI users save at least a full workday per week. But 66% get limited or no guidance on what to do with the time saved (BCG).
- More high-value work. 66% of AI users say AI lets them spend more time on high-value work, and 58% produce work they couldn’t have produced a year ago (Microsoft 2026 Work Trend Index).
- Employees feel it. 82% of employees say AI has increased their productivity (Accenture), and 80% of McKinsey’s respondents say AI improved individual productivity (McKinsey).
- Breadth of use multiplies results. 45% of employees who use AI for one or two purposes report a positive productivity effect, rising to 90% of those who use it for seven or more (Gallup).
- The organization matters more than the individual. Culture, manager support and talent practices account for 67% of reported AI impact, more than twice the share explained by individual mindset (Microsoft).
- Managers move the needle. Employees who use AI frequently and have managers who actively support it are engaged at a rate 14 points above the U.S. average (Gallup).
- Productivity at the firm level. The most AI-exposed companies grew labor productivity 163% between 2018 and 2025 on average (PwC).
AI Agent Statistics
- Large companies lead. 40% of organizations with more than $1 billion in revenue are scaling AI agents, up from 27%, while smaller organizations stayed flat at 22% (McKinsey).
- Plans outpace governance. 74% of leaders plan to deploy agentic AI within two years, but only 21% have a mature governance model for autonomous agents (Deloitte).
- Agents are already part of the job. 50% of employees regularly work with AI agents (Accenture), and 47% of frontline AI users spend more time managing and directing AI than doing the work themselves (BCG).
- Fast growth. The number of active agents in Microsoft 365 grew 15-fold year over year (Microsoft).
- A shakeout is coming. Gartner predicts more than 40% of agentic AI projects will be canceled by the end of 2027 due to rising costs, unclear value or weak risk controls. It also expects 15% of day-to-day work decisions to be made autonomously by agents by 2028, up from 0% in 2024 (Gartner).
AI Jobs and Wage Statistics
- AI skills pay. Jobs requiring AI skills carry a 62% wage premium on average, up from 57% a year earlier, and demand for those jobs grew 69% against 9% for the overall job market (PwC 2026 Global AI Jobs Barometer).
- Entry-level jobs are getting harder. AI-exposed entry-level roles in the U.S. are seven times more likely to require skills traditionally expected of senior staff (PwC). Employment for software developers and customer support agents aged 22 to 25 has fallen in recent years while most older groups grew (Stanford AI Index).
- Headcount expectations are rising. 39% of McKinsey’s respondents expect AI to reduce their total headcount next year, up from 32%, though only 14% say AI has contributed to workforce reductions so far (McKinsey).
- Net job creation. The World Economic Forum expects 170 million jobs to be created and 92 million displaced by 2030 (WEF).
- Using AI is becoming part of performance. The Wall Street Journal reports that companies are weighing AI use in hiring, reviews and layoffs, including Accenture’s chief executive saying the firm is exiting employees who can’t adapt (WSJ).
AI Risk, Policy and Shadow AI Statistics
- Policies lag usage. Only 40% of employees say their workplace has a policy or guidance on generative AI (KPMG and University of Melbourne).
- Hidden use is common. 57% of employees hide their use of AI and present AI-generated work as their own, and 66% rely on AI output without checking its accuracy. 56% say they have made mistakes in their work because of AI (KPMG and University of Melbourne).
- Unapproved tools. 43% of workers with access to generative AI admitted using unapproved tools against company policy (Deloitte TrustID via Fortune). More than half of employees without the AI tools they need say they will find alternatives and use them anyway (BCG).
- Responsible-use training is half done. 48% of organizations train employees on the responsible use of AI (Accenture).
- Regulation now expects AI literacy. Article 4 of the EU AI Act requires organizations that provide or deploy AI systems to take measures that support the AI literacy of their staff. The duty has applied since February 2025 and was revised in 2026 (Lewis Silkin).
What the Numbers Mean for Your Company
Read together, these AI statistics tell a consistent story. Buying AI is easy. Getting value from it is a people problem. The companies pulling ahead are doing a few things differently:
They train everyone, not just early adopters
Training is the clearest lever on regular use, and most employees still haven’t had enough of it. See our guide to AI training for employees.
They redesign the work
High performers change workflows instead of layering AI on top of old processes. Start with an AI readiness assessment.
They set rules early
Clear policy cuts shadow AI and mistakes. Use our AI acceptable use policy template.
Managers lead by example
Manager support is the strongest driver of adoption. Our AI Champions Program builds that leadership inside your team.
A note on methodology: surveys of large enterprises (McKinsey, Deloitte, Accenture, NVIDIA) report much higher adoption than surveys of all U.S. businesses (the Census Bureau), because most small firms still don’t use AI. Worker surveys (PwC, Gallup, BCG, KPMG, Microsoft) measure individual use, which is usually ahead of official company programs. When you quote a figure, check whose answers it reflects.
AI Statistics for Business: FAQ
What percentage of companies use AI in 2026?
Among larger organizations, nearly 9 in 10 report regular AI use in at least one function (McKinsey). Across all U.S. businesses, about 20% use AI, rising to 37% of firms with 250 or more employees (U.S. Census Bureau).
Are companies getting ROI from AI?
Some are. 37% of organizations report at least some EBIT impact from AI, but only about 6% get 5% or more of their EBIT from it (McKinsey). Only 25% of executives report widespread, sustained value (Accenture).
How many employees have received AI training?
Fewer than half. 47% of employees globally say they have received AI training (KPMG and University of Melbourne), and access to workplace learning fell to 51% of workers in 2026 (PwC).
Why do most AI projects fail to deliver value?
The research points to people and process, not technology: skill gaps (Deloitte), a learning gap in tools and organizations (MIT NANDA), unredesigned workflows (McKinsey) and too little of the AI budget spent on people (Deloitte’s 93/7 split).
Can I use these statistics in my own presentation?
Yes. Please cite the original source linked next to each figure, and feel free to link to this page as a summary.
Turn AI Spending Into Results
We train your employees to use AI on their real work and help you redesign the workflows that produce ROI.
Schedule a demoRelated reading
Sources
- McKinsey, The state of AI in 2026 (August 2026)
- Deloitte, State of AI in the Enterprise 2026
- Deloitte, 2026 Global Human Capital Trends
- Accenture, Pulse of Change (October 2026)
- PwC, 2026 Global AI Jobs Barometer
- PwC, Global Workforce Hopes and Fears Survey 2026
- BCG, AI at Work 2026 and AI at Work 2025
- NVIDIA, State of AI Report 2026
- Microsoft, 2026 Work Trend Index
- Gartner, agentic AI predictions (June 2025)
- Stanford HAI, 2026 AI Index Report
- World Economic Forum, Future of Jobs Report 2025
- Gallup, AI in the workplace indicator
- KPMG and University of Melbourne, Trust in AI global study 2025
- U.S. Census Bureau, Business Trends and Outlook Survey (May 2026)
- Forbes, Only 3% of employers prioritize building AI skills in-house
- Forbes, The AI training crisis
- Forbes, AI literacy leads learning agendas, but training falls short
- Fortune, Deloitte’s CTO on the 93/7 split
- Fortune, MIT report on generative AI pilots
- The Wall Street Journal, The Boss Has a Message: Use AI or You’re Fired
- DataCamp, State of Data and AI Literacy 2026