AI Outpaces Training — The 78% Problem Nobody Is Solving

TL;DR

• 86% of companies plan to increase AI spending, but only 43% invest in upskilling — a 2:1 bet on technology over humans that's systematically failing • Only 24% of employees have full access to AI tools at work, with 68% of organisations struggling with insufficient AI skills • Top 18% of companies ("Talent Reinventors") see 1.8% higher revenue growth by aligning talent strategy with AI investment • Co-creation is key: 56% of leaders and 48% of workers want employees involved in designing AI workflows, not receiving top-down mandates

Eighty-six per cent of companies plan to increase their AI spend this year. Only forty-three per cent plan to upskill the people who will use it. That is not a rounding error. That is a 2:1 bet on technology over humans — and it is losing. The talent reinventors already know this.

We have seen this film before. Cloud migrations that outpaced cloud skills. Digital transformations that digital teams could not sustain. The pattern is always the same: leadership buys the tool, forgets the team, and wonders why nothing changes — a pattern we've documented in why AI projects fail. AI is the same story, except the gap is wider and the stakes are higher.

The numbers that should stop you

Let us start with the data. Accenture surveyed 1,320 C-suite executives and 4,560 employees across 20 industries and 12 countries for their 2026 Talent Reinventors Report. The findings paint a clear picture of an investment mismatch.

86% of organisations plan to increase AI spending. That is nearly nine in ten companies pouring more money into models, agents, and infrastructure. But only 43% plan to invest in upskilling their workforce to use any of it. The gap is not subtle. It is structural.

Then there is the access problem. Only 24% of employees have full access to AI tools at work. Most are watching from the sidelines while a small group experiments. Worse, only 21% feel they have a voice in how AI gets implemented in their organisation. People are not resisting AI. They are being excluded from it.

The Conference Board's research on transforming organisations for AI confirms the same pattern from the employee side. 68% of organisations struggle with insufficient employee skills for AI. More than two-thirds admit their people cannot do what is being asked of them. Yet the training pipeline is a trickle.

Here is the sharpest number: only 9% of workers agree that HR provides effective AI training, compared to 27% of leaders who believe the same. Leaders think training works. Workers know it does not. That 18-percentage-point perception gap is where AI strategies go to die.

Why more tools will not fix this

The instinct in most organisations is to push more tools. If adoption is low, the thinking goes, people just need better tools or more of them. This is backwards.

The data shows the real barriers are not technological. They are human. 55% of employees experience cognitive overload from the pace of change. 49% feel anxious about job displacement. You cannot train people who are overwhelmed and scared. You have to address the overload first, then the skills.

This is what Accenture calls the "productivity paradox" of AI adoption. The technology is ready. The people are not. And the gap between technology investment and human investment is widening, not closing.

Consider the math. If your organisation spends £500,000 on AI licences, AI infrastructure, and AI consulting — but allocates £50,000 to training — you are not under-investing in training. You are actively choosing to fail at adoption. The AI works. The people cannot use it. The return is zero.

What "good" looks like

Accenture identified a group they call "Talent Reinventors" — the top 18% of organisations that align their talent strategy with their AI strategy. These companies do not just buy AI. They prepare their people for it.

The results are dramatic. Reinventors see 1.8% higher revenue growth and 1.4% higher profit growth than their peers. They are 7x more likely to strengthen company culture and report 6x better employee experience. Turnover drops by 40%. Workforce adaptability is 4x higher. Innovation skills see an 11% uplift.

These are not small differences. They are transformational. And they come from treating people as the primary investment, not the secondary one.

What do Reinventors actually do differently? Six things separate them from everyone else:

Clarity. They communicate a clear AI strategy that connects to business outcomes. Employees know why AI matters, not just what it is — clarity that aligns with a mature AI SDLC maturity framework.

Intelligent teaming. They design workflows where humans and AI complement each other, rather than replacing one with the other.

Talent mobility. They move people into new roles as AI changes old ones, rather than making roles redundant and hiring from outside.

Co-learning. They learn alongside their teams. 96% of Reinventors have an integrated talent-technology strategy, compared to just 16% of other organisations. That integration means AI training is not a separate programme — it is embedded in the work itself.

Breakthrough leadership. Leaders at Reinventor organisations model AI adoption themselves. 32% of executives at Reinventor firms use AI daily, up from 8% in 2024. If the CEO will not use it, why should anyone else? This is where fractional CTO services can make the difference — bringing AI-fluent leadership into organisations that lack it.

Personalised experiences. They tailor AI tools and training to individual roles, rather than rolling out a one-size-fits-all platform and hoping for the best.

The co-creation gap

There is one more data point worth highlighting. The Conference Board found that 56% of leaders and 48% of workers believe employees should co-create how work gets redesigned with AI. Both sides want the same thing: involvement. Yet most organisations design AI implementations top-down, hand them to employees as fait accompli, and then act surprised when adoption lags.

Co-creation is not a nice-to-have. It is the mechanism that closes the perception gap. When employees help design how AI changes their work, two things happen. First, the design is better because it reflects actual workflows rather than theoretical ones. Second, adoption accelerates because people support what they helped build.

The Conference Board's framework is straightforward: top-down vision combined with bottom-up innovation, connected through governance and shared learning. Leaders set the direction. Employees identify the highest-value tasks to transform. The interlock happens through structured feedback loops, not mandating tools from on high.

What this means for your organisation

If you are a CEO or founder at a 50-to-200-person B2B company, this data is speaking directly to you. You probably have an AI budget. You probably do not have a training budget that matches it. And you probably have a gap between what your leadership team thinks is happening with AI and what your employees actually experience.

Here are the practical steps:

Audit the ratio. Add up what you spend on AI technology — licences, infrastructure, consulting. Then add up what you spend on AI training, upskilling, and change management. If the ratio is worse than 1:1, you have a problem. Fix it before you buy another tool — and consider AI-native engineering support that embeds training into the engineering process.

Close the perception gap. Survey your team. Ask two questions: "Do you have the skills to use AI effectively?" and "Does our training prepare you for AI?" If your employees give different answers than your leadership team, you have a blind spot.

Give people a voice. Before you roll out the next AI initiative, ask the people who will use it what they need. 56% of leaders and 48% of workers want co-creation. Deliver it.

Embed learning in the work. Standalone training programmes have diminishing returns. The Reinventors integrate learning into daily workflows. AI training should happen at the point of use, not in a separate session that people forget by Friday.

Model it from the top. If your executive team is not using AI daily, your organisation will not either. 32% of execs at leading firms use AI daily. That is not a coincidence. It is a prerequisite.

Key Takeaways

• The 2:1 investment mismatch is systematic: 86% of companies increase AI spending while only 43% invest proportionally in training, creating predictable adoption failures across industries

• Access and voice drive adoption more than tools: Only 24% of employees have full AI access and just 21% feel heard in implementation decisions — exclusion, not resistance, is the real barrier

• Talent Reinventors prove ROI comes from people-first strategies: The top 18% of companies see 1.8% higher revenue growth and 40% lower turnover by aligning talent and technology investments

• Co-creation closes the 18-point perception gap: When employees help design AI workflows rather than receiving top-down mandates, both adoption and effectiveness accelerate dramatically

• Leadership modelling is non-negotiable: 32% of executives at successful AI companies use AI daily versus 8% previously — if the C-suite won't adopt, nobody else will either

The bottom line

The AI adoption problem is not an AI problem. It is a training investment problem. It is a people problem. It is a "we bought the tool and forgot the team" problem.

The companies getting AI right — the Reinventors, the top 18% — do not spend less on technology. They spend proportionally more on humans. And they get better returns because of it: higher revenue, higher profit, lower turnover, stronger culture, faster innovation.

You do not need more AI tools. You need more people who can use the ones you have. Start there.

Ready to close the gap? Stop spending on AI tools your team cannot use. Invest in readiness first. Book a people-first AI readiness session with TechLevity — we will map your current adoption, identify the perception gaps, and build a training plan that matches your technology investment.

[Book your AI readiness session →]

<!-- Internal linking opportunities:

• AI strategy implementation: Link to /insights/why-ai-projects-fail and /services/ai-strategy
• Leadership AI adoption: Link to /services/fractional-cto for executive guidance
• AI governance frameworks: Link to /insights/shadow-ai-governance-guide
• Training program development: Link to /insights/ai-sdlc-maturity-framework
• Related failure patterns: Link to /insights/200k-ai-pilot-never-shipped
• Organisational readiness: Link to /insights/agent-architecture-production-ai

-->