ZipRecruiter AI Worker Report: AI skill job postings up 159%
The ZipRecruiter AI Worker Report, released on October 26, 2026, found that job postings calling for AI capabilities on ZipRecruiter rose 159% from a year earlier, while only 29% of 1,500 surveyed U.S. workers said they receive enough AI training. Senior leaders report far more benefit than entry-level staff. For employers, the report points to a gap between what they ask of workers and the training they give them.
Why employers are asking for AI skills
Demand for AI skills in job listings is rising fast. As of September 2026, ZipRecruiter's marketplace data shows that postings calling for AI capabilities are up 159% from a year earlier. Such postings now make up over 4% of all listings, compared with less than 2% a year ago.
ZipRecruiter chief executive Ian Siegel said experience working with AI is the fastest-rising job requirement on the marketplace, for both technical and non-technical roles.
"We are seeing a seismic shift in the ZipRecruiter job search marketplace." — Ian Siegel, CEO of ZipRecruiter
What workers told ZipRecruiter about AI at work
Most surveyed workers welcome AI. The report, titled Better Work, More to Learn: The 2026 AI Worker Report, is based on a survey of 1,500 employed U.S. workers. Of them, 53% said AI has made their work more enjoyable, compared with 6% who said it made work less enjoyable.
At the same time, 42% worry that AI will make their job or skills obsolete. Workers can value the tool today and still fear what it means for their careers.
Why senior leaders benefit more than entry-level staff
The gains are concentrated at the top. Some 77% of senior leaders say AI has improved their work satisfaction, compared with 40% of entry-level employees.
ZipRecruiter notes that many tasks AI now assists with are the ones entry-level workers are most often asked to do, such as routine research, documentation and data processing. The survey does not explain the gap. Job duties, access to approved tools, autonomy and training could all play a part, HRTech Edge noted.
How AI training lags behind employer encouragement
Employers push AI use more than they teach it. While 39% of workers say their employer actively encourages AI use, only 29% say they receive enough training, and 37% are learning the tools on their own.
AI is also raising expectations. Half of workers, 50%, report higher productivity expectations, and 45% say AI is making technical skills outdated faster. Without training and realistic workload planning, those expectations can shift pressure onto employees instead of reducing their work.
What the report means for HR teams and retention
The findings carry a retention risk. One in four respondents, 25%, plan to leave their jobs because of AI-related concerns, and 31% want their next role to involve less AI. The survey does not show that they will actually leave, but stated intentions on that scale are a signal for workforce planners.
HRTech Edge suggests practical steps for HR and learning teams: training tied to each role's real tasks, guidance on confidential information, rules for checking AI output and clear escalation when the tools make errors. It also recommends explaining how AI will change specific jobs and which responsibilities will stay with people, and defining the AI skills each role needs rather than treating AI proficiency as one universal requirement.
Questions and answers
How fast is demand for AI skills growing in job postings?
As of September 2026, job postings on ZipRecruiter that call for AI capabilities were up 159% from a year earlier and made up over 4% of all listings, up from less than 2%.
Do workers like using AI at work?
Mostly yes. In ZipRecruiter's survey, 53% said AI made their work more enjoyable and 6% said the opposite, though 42% worry AI could make their job or skills obsolete.
Are employers training workers on AI?
Not enough, by workers' own account. 39% say their employer encourages AI use, but only 29% say they get enough training, and 37% are teaching themselves.
MOZGII view
How we would build it
The gap in this report is a training system that does not exist yet in most companies. Employers ask for AI skills in job postings, encourage staff to use the tools, and then leave 37% of workers to learn on their own. The people losing out are entry-level staff, whose routine tasks are the ones AI touches first.
If we were building a training system for this, we would start from each role's real tasks, not from a general AI course. For a support agent, that means drafting replies and summarizing tickets; for an analyst, cleaning data and checking a model's numbers. Each lesson would use the company's approved tools and its own documents, and a manager would review a sample of each learner's AI-assisted work before signing off the skill. Rules on confidential data and on checking output would sit inside every lesson, not in a separate policy nobody reads.
It would break where managers have no time to review work, and where the approved tools lag behind the ones staff use anyway. The one metric we would watch is the share of entry-level employees who complete role-based training within their first months, set against how satisfied they report being with AI, because the report shows that is where the gap is widest.
Next, we would build a skills record that lists the AI tasks each employee can do, so internal hiring can use it instead of guessing.
Sources
Where the facts in this story come from.