Skillsoft research has found that AI use at work is now widespread, but only 24% of employees feel fully equipped with the skills needed to use the technology effectively.
The company’s Workforce Readiness Report: AI Edition found that 86% of surveyed employees use AI tools at work. The gap between usage and confidence suggests that AI adoption is moving faster than formal training, governance, and role design inside many organisations.
Employers are investing in AI tools, pilots, and automation projects, while many employees are left to interpret new systems through informal experimentation, peer advice, vendor prompts, or unapproved tools. That can lift short-term productivity while creating inconsistent quality, data risk, and uneven capability across teams.
AI is already embedded in many working routines. The harder questions sit around whether companies have built the skills, operating models, and management practices needed to use it safely and productively at scale.
Generative AI can appear deceptively simple. Employees can produce summaries, drafts, code snippets, research notes, slide outlines, and analysis quickly. Effective use still depends on judgement: knowing what data can be entered, how outputs should be checked, when human review is essential, where bias may appear, and how automated work changes accountability.
Government interest in AI’s effect on jobs, skills, productivity, and work quality has already placed the issue inside the policy debate. Skillsoft’s research adds company-level evidence to that question. Employees are using AI, but many do not feel prepared, leaving productivity gains uneven and harder to measure.
The gap complicates leadership assumptions about return on investment. Boards may approve AI spending on the basis of efficiency, automation, and faster decision-making. Those benefits depend on adoption quality. A workforce that uses AI without adequate training may produce more output, but it may also require more review, increase compliance exposure, or generate errors that are harder to detect because they appear polished.
Managers are central to the transition. They need to know which tasks should be automated, which should be assisted, and which should remain human-led. They also need to assess whether productivity gains are being captured as better service, faster delivery, improved quality, or simply higher workloads.
Training programmes therefore need to move beyond generic AI awareness. Employees in finance, HR, legal, marketing, customer service, operations, sales, and software teams face different risks and opportunities. A single prompt-writing course will not solve role-specific questions around data governance, decision rights, quality control, customer trust, and regulatory exposure.
The readiness problem also affects early-career development. Many junior roles are built around tasks that AI can now assist or partially automate: drafting, summarising, analysis, research, reporting, coding support, and document preparation. If those tasks disappear without alternative training pathways, companies may weaken the pipeline of future managers and specialists.
Workforce confidence will influence adoption. Employees who fear being replaced may resist using AI openly. Others may exaggerate competence to avoid being seen as behind. Both behaviours make it harder for employers to assess true capability and design appropriate support.
Companies that move fastest may not be those with the most tools, but those with the clearest operating discipline. Approved use cases, role-level training, secure systems, quality review, and manager accountability are now part of AI implementation. Without them, AI adoption risks becoming a distributed experiment rather than a managed transformation.
Skillsoft’s research places workplace AI firmly inside capability management. The next stage will depend less on whether employees try AI and more on whether organisations can turn experimentation into governed, measurable, and trusted work.
You must be logged in to post a comment.