Closing the AI skills gap: a coaching-led approach

Closing the AI skills gap: a coaching-led approach

The gap between the AI skills companies need and what employees can deliver is widening fast, and it now extends well beyond data scientists to everyone who touches AI tools. According to Skillsoft's 2024 IT Skills and Salary Report, 56 percent of respondents expect skill gaps to persist in their teams over the next one to two years, even as AI investment grows. The core problem isn't a lack of information; it's that habits haven't changed. Coaching-led upskilling closes this gap by building confidence, sharpening judgment and turning knowledge into daily practice.

What is the AI skills gap: and why is it widening?

The AI skills gap is the mismatch between what companies need to stay competitive and what their workforce can currently do, spanning technical depth, practical application and foundational AI literacy across every department.

‍

According to McKinsey's 2024 Global Survey on AI, 72% of companies have adopted AI in at least one business area, yet fewer than half say they have enough talent to support their AI goals. Gartner's 2024 research found that AI talent shortages rank among the top three barriers to enterprise AI rollout worldwide.

‍

Generative AI has made the situation more urgent. Tools that were rare in 2022 are now part of daily work, or at least they should be. In Germany, a 2024 study by the Bitkom Digital Association determined that 68% of companies see AI as strategically important, but only 20% have formal AI skills programs. With the EU AI Act implementing new compliance and literacy rules in 2025, the skills gap is now a regulatory risk as well as a competitive one.

Root causes: why training alone has not closed the gap

Traditional L&D budgets can't keep pace with how quickly AI evolves. By the time most companies launch an AI training program, the tools have already changed. But speed is not the only issue.

‍

Most organizations rely on general, course-based methods. LMS modules, external bootcamps, certification programs. These approaches transfer knowledge but don't change behavior. Someone might complete a prompt engineering course and return to old habits the next day.

‍

Forbes notes that while AI technology advances rapidly, many work processes remain unchanged because professionals need more than efficiency gains to fully adopt new tools. The article also highlights that concentrating AI expertise in dedicated units prevents the broader workforce from developing hands-on capability. BCG's 2024 AI Radar report found that only 6% of companies have moved beyond piloting to deploy AI at full scale, evidence that this isolated-team approach is widespread.

‍

Without ongoing, personalized support, employees may understand an AI tool but still never use it consistently.

The gap beyond data scientists: AI literacy across every role

According to the World Economic Forum's Future of Jobs Report 2025, 77% of employers plan to reskill or upskill their current workforce in AI-related skills by 2027. Most of these roles are non-technical.

‍

A workable framework divides AI skills into three tiers:

‍

AI literacy (all staff): critical thinking about AI outputs, basic prompt writing, data interpretation and understanding AI limits

‍

AI application skills (functional roles): workflow integration, tool selection and setup, process redesign

‍

AI engineering expertise (technical roles): model building, data architecture, systems integration

‍

Most employees need basic AI literacy, yet most training budgets flow toward advanced engineering skills. This mismatch leaves the largest group of workers unsupported, even though their daily use of AI determines whether investments pay off. HR leaders can rebalance by shifting 10 to 20 percent of upskilling spend from advanced courses to practical, on-the-job AI literacy programs. A good starting point: review current training spend by skill tier and redirect resources to where they'll drive the broadest adoption.

‍

The barriers aren't purely technical. Fear of job loss, resistance to new workflows and a lack of psychological safety to experiment all slow adoption. Companies should make change management a priority, with managers leading the way. HR leaders can accelerate this by offering manager training on AI leadership, creating open forums for employees to voice concerns and establishing peer support groups that normalize experimentation. Organizations that invest in coaching for managers create a culture shift that starts at the top.

How coaching closes the AI skills gap at scale

Coaching works where training alone falls short because it is personal, adaptive and ongoing. A course can show employees what an AI tool does; coaching helps them overcome hesitation, redesign their workflows and actually use the tool day after day. Deloitte's 2026 State of AI in the Enterprise report notes that talent remains a major challenge as AI reshapes job roles. Coaching addresses this by giving employees a safe space to build confidence, develop judgment through guided practice and gain hands-on experience that courses rarely provide. Through structured reflection, employees learn when and how to apply AI in their specific roles.

‍

Regular sessions turn one-time experiments into lasting habits.

‍

One-on-one coaching is especially valuable for managers. When managers use AI tools themselves and guide their teams through changes, adoption spreads organically. A Harvard Business Review report notes that generative AI is expected to reshape job roles across sectors. At the literacy level, coaching helps employees overcome resistance; at the application level, it helps leaders redesign workflows. Digital coaching platforms make this support scalable by connecting employees with certified coaches who deliver steady, personalized guidance that traditional LMS systems cannot match.

Frequently asked questions (Build FAQ Section)

What is the AI skills gap?

The AI skills gap is the mismatch between the AI capabilities a company needs and the skills its workforce currently has. It affects every role, not just data scientists or engineers. According to McKinsey's 2024 research, fewer than half of companies that have adopted AI say they have enough talent to support their goals.

Why is there a shortage of AI skills?

AI tools evolve faster than L&D cycles can respond. Education pathways haven't kept up with demand. Most corporate training still focuses on knowledge transfer rather than behavior change. A report from Skillsoft notes that 65% of respondents still experience skill gaps on their teams, with more than half expecting these shortages to continue in the coming years.

How do you close the AI skills gap?

Closing the gap requires a tiered AI literacy model for all employees, coaching-led behavior change that turns knowledge into habits, leaders who model AI adoption and sustained investment in culture and change management. Training alone falls short without the personal, ongoing support that coaching provides.

What skills are needed for AI?

Most employees need AI literacy: critical thinking about AI outputs, basic prompt writing, data interpretation and understanding AI limits. Functional roles need application skills like workflow integration, tool selection and process redesign. Only a small portion of the workforce needs deep engineering expertise. Companies should invest in the broadest tier first, since that group determines whether AI spending translates to real adoption.

From AI investment to AI adoption: next steps for HR leaders

The AI skills gap won't close by adding more courses. It closes when organizations treat adoption as a behavior-change challenge and invest in ongoing, personalized coaching that builds confidence, sharpens judgment and creates lasting habits.

HR leaders can act now. Start by identifying where legacy processes persist despite AI investment. Surface the cultural barriers slowing adoption. Then introduce coaching programs that support every tier of AI skill. CoachHub's AI transformation coaching programs offer a scalable, evidence-based way to make this change measurable and lasting.