Companies are spending enormous amounts of time, energy and money to train their workers to be ready for AI tools, but these efforts are often falling short — or failing outright, recent reports indicated.
In its 2026 AI Readiness Gap report, Docebo found that 85% of employees say they can’t apply the AI training they’ve received to their day-to-day jobs, despite AI literacy and applied skills ranking as the top priority for both employees and learning leaders over the next 12 to 18 months.
Docebo, a learning platform company, also found that 56% of workers are so overwhelmed by what they call “pre-AI” manual tasks that they don’t have time to learn the tools that are designed to save them time. In addition, 78% of respondents say that learning takes place outside the tools they actually use, like Slack or Salesforce, which means that AI training is a distraction instead of a driver of return on investment.
When AI readiness efforts seem to have fallen short, here’s what to do next, according to experts.
Set parameters and guidelines for AI use
To start, companies should have a clearly stated AI policy that outlines what tools are allowed and how they may be used. Without a policy, employees might be experimenting on their own, Melissa Stout, vice president of operations at Milestone, a professional services firm, told HR Dive.
Such experimental use isn’t captured in typical AI adoption tracking, and for highly regulated industries, like finance and healthcare, employees might be putting customers’ personal identification information into public AI tools.
But aside from avoiding a compliance disaster, having a policy with guidelines on how to use AI can help employees adopt it, Stout said. “If there’s no guidance at all, there’s no collaboration around it, then the minute that it feels too hard or they get the wrong answer, people are going to default back to their normal,” she continued.
Giving workers places to collaborate and discuss tricky issues can also help workers adopt AI in ways that will actually improve their productivity. Milestone has a Slack channel for AI wins, for example. Having these spaces to discuss AI “demystifies it and lets them know it’s OK to talk about it,” Stout said.
Address employee AI concerns and different adoption rates
AI readiness training may also assume that every employee has the same basic knowledge, understanding and acceptance of AI. As with any new technology, people from different demographics and different backgrounds will have different baseline expectations when told to use it, Stout said.
Employees may also be reading news reports about layoffs attributed to AI and become worried that they’re being told to train a technology that will eventually replace them; they may also have concerns about the environmental impacts of the technology, Stout said. Employees will then have different levels of comfort in adopting AI, which can mean some people don’t use it or are put onto teams that become gridlocked because of different adoption rates.
These frictions aren’t the fault of employees, said Rema Lolas, founder and CEO of Grozaic, a team-building platform. Instead, poor change management creates a disconnect between “an organization making a really large investment and wanting things to go really fast,” and the people expected to use it, she said. “That doesn’t flow downstream, and people don’t necessarily know what they’re doing.”
Build a timeline instead of a one shot approach
Teams tasked with AI adoption can be stuck in the middle of C-suite executives who want quick ROI for the money they’re spending on AI tools and workers who are told they need to change the way they work right now or they’ll be shown the door.
“You can’t just send all employees on a one-hour AI training course,” said Megan Beane Torres, vice president of employee success at Docebo. Companies may have also been oversold on the promise of AI, which puts unrealistic expectations on adoption and productivity.
Teams need to be asking questions about AI use, she said.
“What is the problem we had in the beginning that AI solves?” she said. “Let’s not just throw AI at everything.”
Instead of a one-shot approach, learning and development professionals can instead create a roadmap for a “learning journey,” she said, and explain “what each step of that journey is.” If AI readiness efforts have been floundering, that means start with an introduction to AI, including what “A” and “I” stand for. “As you start to dive deeper, that’s when you’re involving business leader pain points and adding personalization by department.”
