Blogs

Chat with us

AI Upskilling in UK Workplaces: 8 Barriers and Who Leads Change

In boardrooms, HR departments, and remote workspaces across the UK, the rise of artificial intelligence is no longer on the horizon. It is woven into daily business activity, changing how decisions are made, and how growth is managed. At EWS Limited, we have seen firsthand how companies adapt, struggle, and succeed with AI tools by their side.

Yet, through new data and our direct experience, one fact is clear: while more than half of UK employers expect AI and automation to change workforce skills in just three to five years, the path to these new skills is full of real, sometimes overlooked barriers (insights on AI impact).

Change is coming, but progress is not guaranteed.

Why AI upskilling is on every agenda

Before addressing barriers and leadership in AI upskilling, it is helpful to understand why this skill shift in the UK is so pressing. In our client conversations and sector reviews, the question is no longer should businesses upskill for AI, but rather, how fast, and who should take responsibility.

Recent surveys support this rapidly growing urgency. Over half of UK employers—52%—report they expect at least some shift in employee capabilities because of AI and automation in the next three to five years (Office for National Statistics). A matching 50% of employers put these technologies at the top of factors behind such change.

AI is no longer an isolated innovation kept within IT teams. One in six UK companies is already using it daily to solve core business problems. For those already on board, workers see the real benefit—77% of them say AI saves them at least an hour a day, and many report saving three or more hours.

Despite this, upskilling remains uneven. Expectations are high, but day-to-day realities pull efforts in different directions.

The opportunity and the paradox: rapid adoption, slow upskilling

If AI brings so much value, why are skill efforts lagging? In our work with UK companies facing international expansion, recruitment, and digital transformation, we often hear the same concerns echoed in national data.

  • Time—both in terms of planning training and attending sessions—is called out as the top barrier by employers (40%) and jobseekers (33%).
  • Responsibility is unclear: 56% of jobseekers feel it is up to them to skill up, while 56% of employers believe senior leadership should set the pace.
  • Cost of training, closely following time as a stumbling block, is cited almost equally by both groups.

The result? Rapid adoption of certain AI tools or processes, but fragmented learning and patchy knowledge. When no clear structure or leader emerges, AI can never move from experimental side project to the heart of business workflow.

Upskilling alone is not enough. Teams need direction, support, and space to learn together.

Eight main barriers to AI upskilling in UK businesses

Drawing on studies from the UK government, sector overviews, and EWS Limited’s direct consulting, the following are the eight most common obstacles that must be addressed if AI is to truly change UK business for the better.

  1. Lack of time for structured learning
  2. Time is, by far, the most cited challenge. One in three jobseekers and 40% of employers say upskilling is squeezed out by normal workloads. Even when funding or software is available, finding uninterrupted hours for self-study, group training, or experimenting with new AI tools is seen as nearly impossible (Flexible AI Upskilling Fund evaluation).
  3. Unclear ownership and leadership
  4. Who is in charge of upskilling? Data shows jobseekers and employers both feel unsure, with more than half looking in different directions for leadership. This hands-off attitude means new AI skills are sometimes left to chance, or training comes too late to make a real impact.
  5. Insufficient digital literacy
  6. Foundational digital skills are still lacking in many workplaces. Without baseline digital literacy—such as basic file management, typing, or password control—even introductory AI training may not stick. The UK government’s own analysis highlights this barrier, noting particular challenges among older adults or those from lower-income backgrounds, who may lack device access entirely.
  7. Existing workload and competing priorities
  8. Even staff who want to upskill struggle to balance daily duties with learning. Pressure to hit KPIs and manage client expectations often leaves little energy or time for AI experiments, peer learning, or training. This is especially true in fast-scaling companies where every day matters.
  9. Training costs and budget constraints
  10. Funding is a real-world barrier, not only for external training but for the “hidden costs” such as time away from work, management oversight, and IT support. Almost half of UK firms surveyed say cost concerns limit their ability to expand AI knowledge across teams (Office for National Statistics findings).
  11. Lack of clear use cases and business understanding
  12. Many companies still cannot identify where AI fits into their specific operations. In 2023, 39% named this as their main adoption barrier. If organizations cannot connect training to real-world applications—how AI supports sales, HR, logistics, or operational risk—learning loses its value and interest stalls.
  13. Uneven sector experiences and digital divides
  14. The UK’s AI readiness is not the same across sectors. Construction lags on basic digital literacy; creative industries embrace tools but sometimes without enough attention to quality or originality (UK government sectoral overview). Health, finance, and IT firms show uneven patterns of rollout and training quality.
  15. Quality and consistency of training
  16. While nearly half of employers support on-the-job training, mentorship, or rotation programs, these methods risk becoming inconsistent or directionless when no one owns the curriculum or outcome. Peer learning is powerful, but only when guided by clear objectives and shared learning milestones.

Business team in a meeting reviewing AI training materials Time: the universal obstacle and why leaders struggle with it

If time is the wall everyone hits, why is it so hard to fix? Time constraints have stubborn roots—busy teams scarce on resources, a rush to market for growing companies, and the sheer pace of digital change.

Matt Burney, an influential UK thought leader on digital workplaces, captures the challenge: “Employers want to move at pace with AI but struggle to provide space or decide who drives the change.” We see his point echoed often: when teams are already strained, new training feels like an extra burden.

Protected time is non-negotiable. Without it, every AI change meets resistance.

His insights resonate deeply with what we see with EWS Limited clients: Productivity gains are not about heaping tools on overstretched teams, but about redesigning work and using protected time for problem-solving and experimentation.

When AI education is tacked on, or when upskilling is left to self-motivation, friction grows. Workloads stay high, pressure rises, and staff adopt AI in patchy or ad hoc ways.

Often, this drives a rift between employees hoping for personal development and leaders expecting quick returns without long-term planning.

The split: whose job is AI upskilling anyway?

This is not a small detail. In practice, it shapes whether an organization thrives or simply muddles through change.

Among employers, 56% expect senior leadership to set both the pace and tone for AI skill development. In our experience, where leaders step up, rollout improves—whether in tech hiring, HR, or global growth projects. For jobseekers or employees, though, a matching 56% feel the task is their own personal responsibility. The confusion means teams rarely have a shared roadmap or mutual accountability.

This is not only a UK challenge. In every market where we deliver skills support or help build global workforce solutions, we see upskilling falter unless it is part of the business strategy, not a side initiative.

When responsibility is muddled, the benefits of AI stay out of reach.

Real approaches to effective AI upskilling

We believe that AI adoption only works when learning stops being a box-ticking exercise. Organizations that create protected time, encourage open experimentation, and offer collaboration have far more positive AI outcomes.

Key recommendations based on both sector data and our work at EWS Limited include:

  • Schedule protected learning time. This includes freeing up calendars, as well as shielding learners from day-to-day interruptions.
  • Develop clear learning journeys. Tie every AI module or training back to business priorities and job roles, especially for managers, global mobility leads, or IT teams.
  • Appoint champions. Effective upskilling needs people or teams who hold both the vision and the responsibility for leading others.
  • Balance training with hands-on, on-the-job AI applications. Encourage mentorship and peer feedback, but subject them to real review and measurable outcomes.
  • Invest in foundational skills. Ensure digital literacy for all. Many workers struggle with basic IT hurdles (barriers to AI skills development).
  • Encourage sector-wide collaboration. Partner with consultancies, like EWS Limited, that understand sector differences and international workforce norms.

Team of professionals attending an AI skills workshop AI adoption rates: lessons from across UK sectors

Sector readiness varies. In IT, finance, and digital-heavy industries, daily use of AI is almost routine. In other fields—construction, logistics, creative, or healthcare—progress is mixed. According to the Office for National Statistics, many companies still cannot spot opportunities where AI adds value to existing workflows. Nearly 39% say identifying these activities remains their primary roadblock.

We have found that companies investing in international expansion face extra hurdles—laws differ, local practices change, and digital maturity is never uniform. Across 100+ countries where EWS Limited delivers Employer of Record and payroll solutions, we see that a scalable HR strategy for global teams needs both digital skills and clarity on who leads transformation.

In creative industries, risk grows if training is informal and unmonitored, leading to mistakes or shots at “AI shortcuts” that hurt brand reputation (UK government sectoral overview). In construction and manufacturing, investing in digital basics is the first step.

No matter the sector, without structure and responsibility, AI learning will be scattershot—missing the billions in time-saving and workflow improvement found in the best-adopting organizations.

“Productivity comes not from new tools, but new ways of working.” – Matt Burney

Case stories: when upskilling works and when it falls short

%0LS����

  • share on Facebook
  • share on Twitter
  • share on LinkedIn

Related Blogs