60% of employees fear AI is weakening critical workplace skills, IBM study finds

A majority of employees believe artificial intelligence may be weakening the very skills organisations increasingly need, according to new research from the IBM Institute for Business Value.

The study found that 60% of employees are concerned AI is contributing to skills erosion, while organisations continue to place growing importance on human capabilities such as critical thinking, judgement and problem framing.

Based on surveys of 8,800 full-time employees and 1,500 Chief Human Resources Officers (CHROs) and senior workforce executives worldwide, the research points to an emerging challenge for employers balancing AI adoption with workforce capability development.

Growing concerns over AI-driven deskilling

According to IBM, concerns about skills erosion extend beyond employees and are increasingly shared by business leaders.

Among employees who worry AI is weakening workplace skills, three in four said the technology has already begun eroding at least some of their capabilities, with critical thinking identified as the most commonly affected area.

The study found:

• 60% of employees are concerned AI is weakening their skills
• 75% of concerned employees say skills erosion has already begun
• 46% of CHROs identify skills erosion as a major concern
• 80% of organisations have established reskilling roadmaps for AI adoption

IBM distinguishes skills erosion from skills gaps. While reskilling focuses on developing new capabilities, skills erosion occurs when employees stop practising existing competencies because AI increasingly performs those tasks.

Critical thinking emerges as a strategic capability

The research highlights a striking disconnect between what organisations value and how employees perceive essential skills in an AI-enabled workplace.

Nearly half of employees (49%) and 57% of CHROs identified critical thinking and problem framing among the most important capabilities for the AI era.

At the same time, 71% of CHROs said the ability to supervise, validate and override AI outputs is the workforce's most essential skill, compared with only 29% of employees who ranked judgement as a top priority.

One of the report's key findings is that as AI becomes more capable of performing cognitive tasks, human oversight becomes increasingly important rather than less relevant.

Human oversight remains a challenge

IBM's research suggests organisations cannot assume meaningful oversight exists simply because humans remain involved in AI-assisted decisions.

The study found:

• 62% of CHROs report rising employee confidence in AI-enabled decisions when human judgement is deliberately embedded into workflows
• 57% of CHROs report declining confidence where judgement is not formally incorporated
• 41% of CHROs believe employees may not feel safe challenging or overriding AI recommendations
• 43% of employees say they are likely to be blamed when AI-related decisions go wrong
• 36% of CHROs cite unclear accountability as a barrier to AI deployment

According to IBM, organisations where CHROs share responsibility for determining which decisions remain human-led report significantly stronger employee confidence in questioning AI-generated outputs.

AI adoption creates new forms of work

The report also challenges assumptions around AI-driven productivity gains.

Four in five CHROs believe AI creates "invisible work", including validating outputs, correcting errors, supplying context and managing exceptions.

Meanwhile, 42% of employees said AI increases their workload or creates additional tasks that often go unrecognised.

The findings suggest productivity benefits may be more complex than simple time savings, particularly when employees must spend additional effort reviewing, correcting and validating AI-generated work.

Leadership and governance gaps remain

The study identifies significant organisational challenges around AI governance and workforce planning.

Key findings include:

• 46% of organisations do not involve CHROs when defining AI strategy
• 73% of CHROs struggle to coordinate consistently across the C-suite
• Only 28% of CHROs report having a joint AI roadmap with IT teams
• Only 26% of organisations clearly define tasks as human-led, AI-assisted or AI-executed
• Organisations that clearly define those roles report an 18% reduction in risk and a 20% improvement in quality

IBM also found 52% of employees have experienced changes to their day-to-day responsibilities during the past year as AI reshapes workplace tasks.

Future workforce development comes into focus

The research raises broader questions about how organisations will develop future talent if AI increasingly performs tasks traditionally used to build experience and judgement.

Nickle LaMoreaux, Senior Vice President and Chief Human Resources Officer at IBM, said AI is changing both how work is completed and where people can create the greatest value. She noted that uniquely human capabilities become increasingly important as AI takes over routine and process-driven activities.

The findings suggest businesses may need to focus not only on helping employees use AI effectively, but also on preserving the critical thinking, judgement and decision-making skills required to supervise and challenge AI systems. As organisations accelerate AI adoption, maintaining those human capabilities could become a central workforce priority.