Business Reporter – Human Resources

AI tools have reshaped the workforce, offering businesses solutions across productivity, marketing, HR, IT and learning. Across the UK, teams are chasing the ROI of AI adoption through higher productivity and lower costs. Whilst the most AI-fluent companies have achieved productivity gains of 163%, many others have found, sorely, that their AI projects haven’t delivered the expected benefits.

 

The cause is becoming clear. Many workers feel unprepared for this shift, and half of London businesses say their workforce doesn’t currently have the skills needed to meet their organisation’s AI requirements.

 

Globally, the AI skills gap has been valued at $5.5 trillion in unrealised productivity. That figure doesn’t just capture the technical IT skills needed for back-end operations. It reflects everyday capabilities such as prompt engineering, critically evaluating AI outputs, digital literacy and workflow integration. These skills are becoming essential at every level, from entry-level to the C-suite.  

 

Increasing digital skills seems like an obvious next step, what is less obvious is implementation. Building an AI-ready team that can capture the ROI gains may be just as much of a strategy game as scaling the tools your company is adopting.

 

It starts with a job description

Investment in digital skills training is strikingly low. Only 7% of UK businesses are mandating that all staff receive annual digital skills training. This is despite the growing prevalence of shadow AI usage, with more than half of all UK workers admitting to using unapproved AI tools independently for daily tasks, often without formal company implementation, which carries its own risks.

 

Training is a step in the right direction, but it can easily become a one and done exercise, without providing a real understanding of how an organisation should use AI. That creates an even bigger hurdle for organisations trying to prepare people for roles that are changing fast. As AI adoption accelerates, job descriptions as we know them are breaking down. Roles now require employees to combine technical AI capability with strategic thinking, business understanding and customer experience.

 

Traditional job descriptions are built from a mix of soft and hard skills: problem-solving, time management, IT proficiency, data analysis. AI now streamlines many of these, making soft skills more efficient and hard skills more accurate. It makes sense that 1 in 4 workers worry that AI could replace them.

 

In actuality, the labour market is splitting into two tiers, a professionalised tier built on blended roles that need human judgement, and a democratised tier of tasks AI now handles outright.

 

The professionalised tier runs on the understanding that AI is only ever as intelligent as the human guiding it. Increasingly, workers need practical AI literacy, data confidence, prompt writing, and the ability to critically evaluate AI outputs, alongside judgement the technology itself can’t supply. When those expectations aren’t reflected in job descriptions, workers show up underprepared and employers end up hiring the wrong people.

 

For employers, this starts at the point of writing the role. A blended job description should name the AI tools the person will use, then spell out the technical and analytical skills needed to use them effectively. This moves skills like critical thinking from generic bullet points into specific expectations tied to those tools.

 

Instead of asking for “AI experience,” employers should define the capabilities they need, such as analysing AI-generated insights, validating outputs, automating workflows, or using AI to improve decision-making.

 

Get that right, and employers hire people who can capture productivity gains rather than simply operate AI.

 

Upskilling the workforce

Beyond redesigning roles, employers need a strategy to close the skills gap instead of relying on one-off training sessions. Teams need to be routinely audited on their AI capabilities, and those capabilities mapped against what each role genuinely requires in practice.

 

Role-specific training may matter more than generic upskilling, as AI’s function differs across roles. Marketing teams may use it for data analysis, operations teams may utilise dashboards, in sales AI is being used for CRM efficiency, and across creative roles AI-assisted design tools are supporting designers.

 

PwC’s 2026 Global AI Jobs Barometer found that skills needed for the most AI-exposed jobs are now changing more than twice as fast as for the least AI-exposed jobs. This widens the net upskilling casts, as role-specific technical fluency outpaces basic digital literacy.

 

Employers should also be looking ahead as much as they audit the present. McKinsey’s research found that 92% of companies plan to increase AI investment. If businesses are planning to scale their AI capability faster than they’re building the skills to use it, they will inevitably lose productivity gains. Forecasting the skills a three-to-five-year strategy will demand is now becoming a necessity, not an afterthought.

 

Foundational digital literacy is equally key. A firm knowledge of secure browsing, digital communication ethics, basic data handling, and using collaborative platforms should be baseline for every employee. Leadership shouldn’t be exempt either, as managers and executives need enough digital fluency to make sound decisions, manage digital teams, and set direction from the top on how AI is used.

 

None of this can fit into a single training course or session. Companies need infrastructure to sustain this level of upskilling, and can do so through micro-learning tools, regular knowledge-sharing sessions, peer mentoring, or partnerships with external training providers. Building these into the workforce will aid team members in matching the pace the tools themselves are moving at.

 

Infrastructure also needs to protect the skills AI can’t replicate. Storytelling, taste, creative judgement and critical thinking. These unique human skills deserve equal investment, because at the end of the day AI may automate workflows and documentation, but it cannot lead a team, interpret context with nuance, or fully understand a company’s mission.

 

Building a workforce with the right skills

Wearing many hats inside one role used to be frowned upon. AI is making it the most efficient way to work, giving employees access to skillsets that would once have taken years and several job changes to acquire.

 

Workers increasingly need to build analytical and AI-related skills alongside traditionally human strengths. Moving beyond simply knowing how to use AI tools requires building roles with AI in mind from the outset, and guiding teams to accurately interpret AI-generated insights, ask better questions, verify outputs, and apply them to real business decisions.

 

Closing the skills gap takes more than teaching employees to use AI platforms. It means designing roles with AI in mind from the start and building an infrastructure that helps workers develop the judgment behind effective AI.

 

 

Adam Field is Chief AI Officer at Tungsten Automation

 

Main image courtesy of iStockPhoto.com and Thawatchai Chawong