Table Of Contents

1. The Broad Strokes: AI Use Is Rising — But Unevenly
2. The Seniority Divide: Leaders vs. Frontline Workers
3. The Industry Split: Tech Soars While Others Stall
4. Age, Gender, and the Demographic Fault Lines of AI Adoption
5. Company Size: Big Firms Scale, Small Firms Struggle
6. Why Millions of Workers Are Still Not Using AI
7. The Training Gap: The Root Cause Behind the Numbers
8. What This Means for Business Leaders Right Now
9. Conclusion: The Divide Is a Choice

AI Workforce Demographics: Who Uses AI at Work — and Who Is Being Left Behind

The headline numbers look impressive. Surveys cite anywhere from 21% to 88% of workers using AI, depending on who you ask and what you count. AI is supposedly everywhere — and yet nearly half of the global workforce still reports never touching it.

The truth is that artificial intelligence is not being adopted evenly. It is flowing along fault lines that most organisations haven't stopped to map: seniority levels, industry sectors, age brackets, gender dynamics, and company sizes. These fault lines determine not just who benefits from AI today, but who holds competitive power tomorrow.

For business executives, the question is no longer simply "Is your company using AI?" The sharper question is: "Which parts of your workforce are using it, and what is the cost of the parts that aren't?"

This article breaks down the most current data on AI workforce demographics — drawing from McKinsey, Gallup, Pew Research, BCG, Deloitte, and the OECD — to show exactly where AI adoption is concentrated, where it is absent, and what every business leader should do about the gaps they find.

The Broad Strokes: AI Use Is Rising — But Unevenly {#broad-strokes}

Let's start with the data landscape, because it is genuinely complex. Different surveys produce dramatically different numbers depending on how they define "AI use" and who they survey.

At the organisational level, McKinsey's 2025 State of AI survey found that 88% of organisations report regular AI use in at least one business function, up from 78% the previous year. Stanford's AI Index put organisational adoption at 78% for 2024. These are enterprise-level numbers, telling you that most companies have some AI somewhere.

At the individual worker level, the picture is far more sobering. Pew Research, surveying a nationally representative sample of over 5,000 U.S. workers in September 2025, found that only 21% of workers say at least some of their work is done with AI — up from 16% a year prior. Meanwhile, 65% of American workers still say they don't use AI much or at all in their jobs. Gallup's Q4 2025 workforce panel found that nearly half of U.S. workers (49%) report never using AI in their role.

The reconciliation between these extremes matters for strategy. An organisation can technically "use AI" because one department runs a chatbot, while 80% of employees remain completely untouched by the technology. The gap between organisational adoption and actual workforce penetration is one of the defining business challenges of this decade.

The Seniority Divide: Leaders vs. Frontline Workers {#seniority-divide}

Perhaps no demographic split is more strategically consequential than the gap between leadership and frontline employees. The data is consistent across multiple sources: the higher you sit in an organisation, the more likely you are to use AI regularly.

Gallup's research shows that leaders report substantially higher and more frequent AI use than other employees, and that this separation has grown over time. Among organisations that make AI tools available, 67% of leaders use AI daily or weekly. Among individual contributors, the figure drops to 46%. BCG found an even starker result: more than three-quarters of leaders and managers use generative AI several times a week, while frontline worker usage has stalled at around 51% — a gap BCG calls the "silicon ceiling."

This divergence has a compounding effect. Leaders who use AI regularly make faster, better-informed decisions. They redesign workflows and set strategy with AI-native thinking. Frontline workers who don't use AI remain stuck in pre-AI productivity patterns. Over time, this does not just create an efficiency gap — it creates a structural divide in an organisation's ability to execute.

Gallup's research suggests an important reason for this gap: leaders find AI's utility more immediately obvious, because their work involves decision-making, synthesis, and communication — tasks where AI tools deliver fast, visible results. Individual contributors in execution-heavy roles often don't receive clear signals about how AI applies to their specific work.

The Industry Split: Tech Soars While Others Stall {#industry-split}

AI adoption rates vary dramatically by sector, and the divergence between industries is widening rather than narrowing.

Gallup's industry breakdown from Q4 2025 shows technology at 77% total AI use, with 57% of workers using it frequently and 31% daily. Finance and higher education follow at 63–64% total adoption. Professional services reaches 62%. At the other end, retail sits at just 33% total AI use, while healthcare, government, and manufacturing range between 41–43%.

McKinsey's data confirms that knowledge management, IT, and marketing and sales are the functions where AI has penetrated most deeply. Agentic AI use is most common in technology, media and telecommunications, and healthcare — where specific use cases like service-desk automation and deep research have developed quickly.

For executives in lower-adoption sectors, this data carries a specific message: your industry's lag is not a sign that AI is irrelevant to your work. It is a sign that the competitive window for first-mover advantage is still open. Healthcare organisations using AI for workflow optimisation, manufacturers deploying AI in quality control, and retail operators using AI for inventory and personalisation are already pulling ahead of peers who are waiting for the technology to "mature."

Age, Gender, and the Demographic Fault Lines of AI Adoption {#age-gender}

Academic research and large-scale surveys reveal that age and gender are two of the clearest predictors of AI adoption, and both carry significant business and social implications.

Age: Pew Research's data consistently finds that workers under 50 are among the most likely to use AI in their jobs, while older workers lag behind. Deloitte found that AI adoption declines with age across the board. A Harvard Business School meta-analysis of 18 studies covering 143,008 individuals across 25 countries found meaningful age-related adoption differences. Notably, even among young people, AI use is not uniform: research from Australia found that 70% of those aged 14–17 had used a generative AI tool, compared to 55% of those aged 23–26 — suggesting that the youngest cohort entering the workforce may carry stronger AI fluency than their slightly older peers.

Gender: A Harvard Business School meta-analysis found that women had 22% lower odds of using generative AI than men. Data from the U.S. Federal Reserve's Survey of Consumer Expectations found 50% of men use generative AI tools compared to 37% of women — roughly a 25% gender gap. This gap is most pronounced in the 45+ age group.

However, the trajectory is shifting. Deloitte predicted that women's experimentation with generative AI would equal or exceed men's in the United States by end of 2025, noting that women's adoption rate tripled in just one year — outpacing the 2.2x growth seen among men. The challenge is that the gap is closing in usage while a trust gap persists. Women are more likely to express concerns about AI inaccuracy, data privacy, and the technology's societal impacts. Research from SSIR notes that women's hesitancy is rooted not in risk aversion, but in risk awareness — and that distinction matters when designing AI adoption programmes.

Organisational dynamics compound the issue. LeanIn.Org's research found that men are 23% more likely than women to be encouraged by their managers to use AI (37% of men vs. 30% of women). When encouragement and access are unequal, adoption gaps become self-reinforcing — and the downstream cost is real, because workers with AI skills are commanding substantial wage premiums.

Company Size: Big Firms Scale, Small Firms Struggle {#company-size}

One of the most consistent findings across global research is that larger companies are further ahead in AI adoption and scaling than their smaller counterparts.

McKinsey's 2025 data shows that nearly half of respondents from companies with more than $5 billion in revenue have reached the AI scaling phase, compared to just 29% of those with less than $100 million in revenues. Larger companies are more likely to have hired for AI-specific roles, implemented governance frameworks, and redesigned workflows around AI capabilities.

The OECD confirms that SME AI adoption remains relatively low compared to both digital technology adoption broadly and to larger firms. Key structural barriers include capital constraints, limited technical expertise, and integration costs that large enterprises can more readily absorb.

That said, the gap is beginning to narrow in unexpected ways. U.S. Federal Reserve data shows that by mid-2025, small businesses were actually adopting AI faster than large firms in some metrics — a reversal attributed largely to lower entry costs for SaaS-based AI tools, which dropped from approximately $50 per month in 2019 to $20–30 per month by 2025. Among companies with 10 to 100 employees, adoption jumped from 47% to 68% in a single year.

The most significant barrier for very small businesses (under five employees) is not cost — it is relevance perception. SBA research found that 82% of businesses under five employees cited the belief that AI is "not applicable to their business" as their primary reason for non-adoption. This is a knowledge problem as much as a technology problem, and it points directly to the value of education, peer networks, and hands-on demonstrations.

If you are looking for a peer community and expert guidance to bridge that gap, the Business+AI consulting network connects executives with AI practitioners who understand the practical realities of mid-market and SME contexts.

Why Millions of Workers Are Still Not Using AI {#why-not}

Understanding the non-adopters is as strategically important as understanding the early movers. When Pew Research asked workers who don't currently use AI why they don't, a layered picture emerged.

Among workers who haven't adopted AI at work, 36% acknowledge that at least some of their work could be done with AI — up from 31% the year before. That means a growing proportion of non-users recognise the technology's potential relevance but have not made the shift. The barriers are both practical and psychological.

Gallup's research found that lack of utility is the most common barrier to individual AI use. People don't adopt tools they don't see a clear use for. This is particularly true among individual contributors who haven't received use-case guidance from their managers. For UK organisations specifically, the leading barriers to adoption include difficulty identifying use cases (39%), cost (21%), and lack of AI expertise (16%).

For women in particular, privacy concerns and a rational assessment of AI's opacity play a meaningful role. Research from the Federal Reserve Bank of New York found that when women say they want AI training, they are often signalling awareness of the technology's limitations — not just a skills gap. Trust architecture matters as much as functionality.

For many organisations, the honest answer is that non-adoption reflects a leadership and communication failure as much as an individual reluctance. When managers don't model AI use, when use cases aren't clearly communicated, and when training is unavailable or generic, non-adoption is the rational response.

The Training Gap: The Root Cause Behind the Numbers {#training-gap}

Across virtually every major study, training emerges as the single most powerful lever for closing adoption gaps — and the single most neglected one.

Pew Research found that among workers who had taken training in the past year, only 24% reported any AI-related content. Randstad data shows that only 13% of workers have received AI training in past years. The World Economic Forum estimates that 59% of the global workforce will need reskilling by 2030, while skills in AI-exposed jobs are changing 66% faster than in other roles.

The cost of this gap is measurable. IDC research estimates that sustained AI skills shortages risk $5.5 trillion in lost market performance globally by 2026. On the positive side, the opportunity is equally clear: research found that when employees are actually trained on AI, adoption jumps from 25% to 76% — a three-fold increase.

BCG established a useful benchmark: employees who receive at least five hours of AI training show significantly higher regular usage and confidence. The format matters too — in-person coaching makes the biggest difference for building AI literacy across age groups, particularly for workers who are uncertain about the technology's relevance to their specific role.

Deloitte's 2026 State of AI in the Enterprise report identifies insufficient worker skills as the biggest barrier to integrating AI into existing workflows. Yet most organisations respond with general AI literacy courses rather than role-specific, workflow-integrated training. The companies pulling ahead are those treating AI adoption as a redesign challenge — not just an education challenge.

For executives who want their teams to develop practical, hands-on AI capability, Business+AI's workshops and masterclasses are built specifically around real business applications rather than generic tool tutorials.

What This Means for Business Leaders Right Now {#what-this-means}

The demographic data on AI adoption carries several concrete implications for any executive managing a team, a division, or an entire organisation.

Map your own adoption landscape before benchmarking externally. Most organisations know their headline AI usage statistics but have not mapped adoption by role level, department, age cohort, or gender. Start with an honest internal diagnostic. Where is AI being used daily? Where is it absent? What are the actual barriers in each pocket?

Address the seniority gap deliberately. If leaders are using AI and frontline workers are not, the organisation is capturing only a fraction of the potential productivity benefit. Bridging this gap requires more than making tools available — it requires managers to actively model AI use, communicate specific use cases, and build AI into the rhythm of team work.

Treat the training gap as a strategic risk. Generic AI literacy sessions deliver limited returns. Role-specific, workflow-integrated training — with at least five hours of hands-on practice — is what actually moves adoption numbers. Budget accordingly and measure outcomes.

Build for inclusion, not just adoption. The gender and age gaps in AI use are not just equity issues — they are competitive ones. A workforce where only certain demographic groups use AI fluently is a workforce leaving productivity and innovation on the table. Addressing unequal managerial encouragement, privacy concerns, and accessibility of training is not a "soft" priority — it is a business performance priority.

Think beyond pilots. McKinsey's data shows that only one-third of organisations have moved AI from pilots to scaled deployment. High performers — those capturing the most enterprise value — are three times more likely to have fundamentally redesigned workflows around AI. That is the real competitive differentiator: not which AI tools you have licensed, but how deeply AI has been woven into how your organisation actually works.

If you want to connect with peers who are navigating these exact challenges, the Business+AI Forum brings together executives across industries to share what is actually working — not just what looks good on a slide.

Conclusion: The Divide Is a Choice {#conclusion}

The data is clear: AI adoption is not a uniform wave washing evenly across the workforce. It is highly concentrated — in knowledge roles, at leadership levels, in technology and finance sectors, in large enterprises, and among workers who have received targeted training.

For business leaders, this is simultaneously a warning and an opportunity. The organisations that are moving from pilot to scale, redesigning workflows, and investing in targeted upskilling are not doing so by accident. They have made deliberate choices about where and how to embed AI into the fabric of their operations.

The workers and companies on the wrong side of the adoption gap are not there because AI is irrelevant to them. They are there because no one has yet made a compelling, practical case for why and how AI fits into their specific context. That is a solvable problem — and solving it is where competitive advantage is built in the next three to five years.

Ready to Move Your Organisation from AI Talk to AI Results?

Business+AI is Singapore's leading ecosystem for executives who want to turn AI strategy into business impact. From peer forums and expert consulting to hands-on workshops and masterclasses, we bring together the people, knowledge, and practical tools you need to close your own AI adoption gap.

Explore Business+AI Membership →

Join a growing community of executives, consultants, and solution vendors who are moving beyond the hype — and building real AI-powered businesses.