IBM released a new CHRO study on September 21, 2026, saying AI is changing how work is organized around human oversight rather than only automation. In the study, 71% of CHROs said the ability to supervise, validate, and override AI outputs is the most essential workforce skill, while 29% of employees ranked judgment as important.
The study surveyed 1,500 CHROs and 8,800 employees globally. IBM also reported that 60% of employees worry AI is eroding their skills, with critical thinking cited most often as declining, and that 80% of CHROs said AI adoption creates “invisible” work for employees.
An independent write-up by THE Journal said 52% of employees reported their assigned tasks changed during the past year as AI reshaped day-to-day responsibilities. The supplied sources do not identify what specific tasks changed, beyond that broad description.
What IBM says changed
IBM’s study frames the shift as a reordering of priorities inside organizations. The reported gap is between what CHROs and employees say is most important: IBM said CHROs placed supervising, validating, and overriding AI outputs at the center of the skill set they need, while employees were much less likely to describe judgment as essential.
The supplied sources do not identify which industries, job families, or countries were most affected within the global sample.
What employees said
IBM said 60% of employees worry AI is eroding their skills, and that critical thinking was cited most often as declining. That finding does not by itself say whether the erosion is temporary or permanent, only that workers reported concern about it. THE Journal’s independent reporting adds that more than half of employees said their assigned tasks changed over the past year as AI reshaped daily responsibilities.
The material in the research bundle points to a near-term labor story that is different from a simple headcount narrative. The reported change is about how work is being divided, how much human checking is required, and how much judgment employers want workers to supply around AI systems. The supplied sources do not identify any layoffs, hiring plans, or formal retraining program tied to this study.
What to watch next
The main question raised by the study is whether organizations respond by formalizing human oversight duties, or by continuing to add those duties as “invisible” work. IBM’s findings suggest that AI strategy is already shaping expectations for the workforce, but the supplied sources do not state whether the study itself announced a company action plan.
For workers, the most concrete signal in the report is that judgment and oversight are being elevated even as employees say critical thinking is under strain. For employers, the study suggests a growing mismatch between how AI is deployed and how responsibilities are defined. The evidence provided here stops short of showing which companies have changed policy in response.
THE Journal’s account also makes clear that task redesign is already happening in at least some workplaces, but the supplied sources do not identify a universal timetable for those changes. What they do show is a specific split in perception: IBM’s CHRO respondents are emphasizing supervision and validation, while employees are reporting skill erosion and changing tasks.
Why this matters
This report matters because it treats AI’s labor impact as a change in work design, not only a story about automation. IBM’s findings, as supplied, point to more demand for people who can check and override AI outputs, and more concern among employees about the condition of the skills they rely on to do that work.
The evidence does not support broad claims about the entire labor market, but it does show a concrete organizational pattern: AI is being tied to oversight, accountability, and extra work that workers may not always see as part of the job. That is the labor shift this study puts on the record.