Preparing Your Team to Work with AI: Culture Before the Tool

When senior executives were asked in McKinsey's survey (October and November 2024) what holds back AI deployment, they were more than twice as likely to blame employee readiness as to blame their own role, and 46% of them cited skills gaps. Yet the employees themselves say they use the tool more than those leaders estimate. So who is really lagging? The survey notes that about 81% of respondents were American, so its percentages should not be read as a verdict on every country, but its lesson about the perception gap between the top and the shop floor deserves attention.

What do employees ask for?

In the same survey, 48% of employees said formal training was the thing most likely to increase their daily use of the tool, and more than a fifth reported little or no support. This fits with BCG's June 2025 survey of more than 10,600 workers in 11 countries: 51% of frontline workers use AI regularly, a share that has stalled; 36% of employees feel adequately trained; and 25% of frontline workers say their leaders give enough guidance. Those with five or more hours of training, especially in-person, coached training, were more likely to become regular users.

These are surveys of opinions and habits, not controlled experiments, so they show association, not causation. But they suggest that the tool alone is not enough, and that a few well-designed hours may count for more than long, scattered workshops.

Psychological safety

In 1999 Amy Edmondson published a field study of 51 work teams at a manufacturing company. She defined team psychological safety as "a shared belief held by members of a team that the team is safe for interpersonal risk taking". She found that psychological safety was associated with learning behaviour, and that learning behaviour mediated the link between psychological safety and team performance. The relevance to AI is plain: anyone learning a new tool will make mistakes at first, and a team that fears ridicule or punishment will hide its errors and learn less.

What does the labour market ask for?

The World Economic Forum's 2025 report says that 86% of the employers surveyed expect AI to transform their businesses by 2030, and that 85% plan to make upskilling a priority. Analytical thinking is the most sought-after core skill, with about seven in ten companies calling it essential, and creativity, flexibility, curiosity and lifelong learning are gaining importance. In other words, what is wanted is not mastery of a button in one tool but a mind that questions, reviews and learns.

Steps for the leader

This is a practical synthesis based on the above, not a set of individually proven findings:

• Listen before you train. Ask your team where they are afraid and where they need help, and compare that with what you assume.
• Be candid about your purpose. If the aim is better quality of work, say so; if the plan involves changing roles, say that too, since vagueness feeds fear.
• Start small with visible proof. John Kotter's change model combines a sense of urgency, a coalition to lead the change, the removal of barriers and short-term wins, stages that suit a small pilot everyone can see.
• Train on real tasks for at least five hours, with a coach or an experienced colleague.
• Write the rules of use: what is permitted, what data must never be entered, and who reviews the outputs.
• Learn in front of them. When a leader tries, errs openly and corrects, the team learns that error is a path to learning, not a scandal.

Trust in God does not conflict with taking the means; whoever has done their part in listening, training and safety, and then relies on God, may hope that He blesses their work.