First day cover ·
Most HR Leaders Say Staff Can Learn AI Jobs, Yet Companies Keep Hiring Outside
Most HR leaders think current staff can be retrained for AI roles, but many firms still recruit outsiders. Here's what the gap costs and why.
Emilija Manevska/Getty
Here's the thing that stands out in new research from workforce solutions provider Careerminds: the people who run hiring largely believe their own employees can grow into AI work, and yet they still reach outside the building for those roles. The survey suggests a gap between what companies say about internal talent and what they actually do with it.
Confidence in internal talent is high
In the Careerminds white paper, titled The AI Redeployment Gap: What Companies Lose by Hiring over Reskilling, 92% of HR professionals and 88% of people managers said employees in their organizations could be reskilled for AI-related roles. That's a strong vote of confidence, and it's hard to dismiss as a fringe view.
The behaviour doesn't match. Fifty-one percent of HR professionals and 47% of people managers said their organizations were hiring externally for new AI positions. Only 42% in each group said they were moving existing employees into those roles. Put simply, the belief is there, but the pipeline isn't.
Mark Saddic, senior partner at Careerminds, put the problem bluntly: "Companies are investing in AI while potentially overlooking one of their most valuable resources: the people they already employ." My read is that this is the most useful line in the whole study, because it frames reskilling as an asset-management question rather than a training-budget one.
Why internal moves stall
The biggest obstacle isn't willingness. It's visibility. Forty-three percent of HR professionals and 34% of managers cited limited insight into individual skill sets, and into which employees are good candidates for retraining, as a barrier. If you can't see what your people can already do, an external job posting looks like the safer bet.
This isn't a new problem for AI. Teams in this position typically have plenty of learning content but no reliable map of who knows what. Skills data tends to live in scattered places: performance reviews, certification records, project histories and the memories of a few managers. When a new role opens up, the search defaults to the market because the internal search is slow and uncertain.
The skills-tracking market is responding. In September, education company Pearson agreed to acquire Workera, an AI-native skills platform that the research notes is used by organizations including ServiceNow and Accenture. A deal like that signals that vendors see skills mapping as a product category in its own right. Whether those tools change hiring habits is a separate question, and the Careerminds data doesn't answer it yet.
There's also a perception gap to watch. Eighty-seven percent of HR professionals and 84% of managers said their organizations were adequately considering reskilling for future roles. Those figures sit awkwardly beside the hiring numbers. Organizations may believe they are prioritizing internal development, while their hiring decisions suggest otherwise.
Sector differences are wide
The pattern isn't uniform. In technology and telecom, external hiring outpaces redeployment by 15 percentage points. Those are sectors where AI demand tends to be sharpest and where talent is most expensive to recruit, so the pull toward outside candidates is easy to understand.
Media and entertainment is the outlier. It's the only sector in the survey where redeployment matches or beats external hiring for both groups. Among HR professionals in that industry, 56% reported redeploying existing employees, compared with 25% who said they hired externally. Those numbers suggest that some industries already run internal mobility well, and that other sectors could learn from how they do it.
The cost of hiring outside
External hiring isn't just slower. Separate research from The Josh Bersin Company, cited in the Careerminds white paper, found that external hiring can cost three to five times more than filling comparable roles internally. It also takes up to 20 days longer on average.
For a company trying to build AI capacity quickly, that math is hard to ignore. A redeployed employee already knows the product, the internal tools and the people. An external hire starts from zero on all three, and may arrive with a higher salary expectation. Reskilling is slower to set up but can pay back faster once a program is running.
What to watch next
The study's sample is US-only. It covers 595 HR professionals and 547 people managers, aged 25 to 65, working at organizations with 500 to 10,000 or more employees. Fieldwork ran from May 15 to 26, 2026, and Censuswide carried out the survey. Those details matter, because larger firms with mature HR systems may behave differently from smaller businesses that have no skills inventory at all.
Still, the direction is clear enough to act on. If you manage a team, the practical first step is an honest inventory of what your people already know, followed by a short list of roles where a internal candidate could realistically move with targeted training. Hiring outside should be the answer you reach after that search comes up empty, not the default. Watch for whether companies start treating skills data as seriously as they treat their job boards.
