Learning objective: By the end of this article, you will understand why AI adoption in operations depends on people, not just tools, and how leaders can build the confidence, skills, routines, and guidance teams need to use AI effectively in daily work.
AI adoption starts with people
AI is becoming part of everyday work. It can help people write, summarise, analyse, search, compare, classify, plan, and make faster decisions. In operations, this creates a real opportunity. Teams can reduce repetitive admin, respond faster to issues, understand performance more clearly, and spend more time improving the work instead of chasing information. But there is one point organisations should take seriously: AI does not create value just because the tool is available. People have to know how to use it, when to use it, and why it matters.
That is where people-centred AI in Operational Excellence becomes important. Operational Excellence has always depended on people. Processes matter. Systems matter. Measures matter. But people are the ones who notice problems, use the process, follow the standard, improve the work, and hold the gains. The same is true with AI. A company can introduce the most advanced AI tool in the market, but if employees are unsure, unsupported, or unclear about what is allowed, adoption will stay shallow.
Tools alone do not create value
McKinsey’s 2025 workplace AI report makes this point strongly. It says the biggest barrier to scaling AI is “not employees—who are ready—but leaders,” because leadership is often not steering the change fast enough. The same report also says that almost all companies are investing in AI, but only 1 percent believe they have reached maturity. That is a useful reminder: AI success is not only about investment. It is also about leadership, adoption, skills, and the way work changes in practice.
This matters for operations because operational work is full of daily decisions. People decide how to prioritise requests, how to respond to problems, how to escalate issues, how to communicate with customers, how to review data, and how to improve performance. AI can support many of these tasks, but only when people feel confident using it. If employees feel they are experimenting alone, they will use AI quietly, inconsistently, or not at all.
Many employees are not against AI. In fact, many are curious about it. They can see that it may save time and reduce frustration. But curiosity is not the same as confidence. A person may be willing to use AI, but still wonder: Am I allowed to put this information into the tool? Can I trust the output? What if the answer is wrong? Will my manager think I am cutting corners? Is this going to replace part of my role? These are normal questions, and leaders should answer them clearly.
Confidence comes from clear permission
A people-centred approach starts with permission. Employees should know where AI use is encouraged, where it is restricted, and where it is not appropriate. This does not need to be complicated. In many cases, simple guidance is enough. For example, teams can be told that AI may be used to summarise non-sensitive notes, draft internal messages, generate improvement ideas, compare options, or prepare first drafts. They can also be told not to enter confidential customer data, personal information, trade secrets, or anything that violates company policy.
This kind of clarity removes hesitation. When people know the boundaries, they can work more confidently inside them. Without boundaries, AI becomes either underused or misused. Some people avoid it because they are unsure. Others use it in ways that create risk. Clear guidance helps both groups.
Training should connect to daily work
Training is the next step. AI training does not need to start as a large corporate programme. It can begin with practical examples from daily work. Show people how to write a good prompt. Show them how to ask AI to summarise a long procedure. Show them how to turn messy notes into action items. Show them how to compare root causes, prepare a checklist, or draft a simple process communication. The training should feel relevant to the work people already do.
Reuters reported on a Google-backed AI Works pilot in which UK workers could save an average of 122 hours per year by using AI for administrative tasks. The report also highlighted something very practical: permission, encouragement, and a few hours of training helped increase adoption, especially among workers who had not previously used generative AI at work.
That is an important lesson for Operational Excellence professionals. Sometimes the barrier is not resistance. Sometimes people simply need a safe starting point. They need to see how AI helps them today, not in some distant future. A supervisor may use AI to prepare a clearer shift handover. A process analyst may use it to organise interview notes. A manager may use it to turn performance data into discussion questions. A team member may use it to draft a standard response, then review and improve it before sending.
Connect AI to real operational work
The key is to connect AI to real work. General AI enthusiasm is not enough. People need practical use cases that make their day easier and their work better. In operations, useful examples might include summarising recurring issues, preparing meeting agendas, drafting improvement charters, analysing customer comments, creating training checklists, reviewing procedure clarity, or turning lessons learned into action plans.
AI should support thinking, not replace it
But AI should not become a shortcut around thinking. It should support thinking. This distinction matters. People still need to check the output, apply judgement, and understand the context. AI can suggest root causes, but the team should still validate them. AI can draft a procedure, but the process owner should still review it. AI can summarise feedback, but leaders should still listen to the people doing the work.
This is why AI confidence should include AI judgement. Employees should learn how to question AI outputs. They should ask: Is this accurate? Is this complete? Is this relevant to our process? What evidence supports it? What might be missing? Does this match our policy, customer need, or operational reality? The goal is not blind trust. The goal is informed use.
Leaders need to model responsible use
Leaders also need to model the behaviour they want to see. If managers talk about AI but never use it themselves, teams may treat it as another passing initiative. If leaders use AI openly and responsibly, the message becomes more credible. A manager can say, “I used AI to prepare a first draft of this meeting summary, then I checked and edited it.” That simple statement normalises responsible use. It shows that AI is a support tool, not a secret or a gimmick.
Small routines make AI adoption stick
Routines make adoption stronger. Teams should not rely on occasional enthusiasm. They should build small AI habits into existing operational routines. For example, after a weekly performance meeting, AI can help summarise actions. During a problem-solving session, AI can help organise possible causes. Before updating a procedure, AI can help identify unclear wording. After customer feedback is collected, AI can help group common themes. These are small uses, but they make AI part of the way work improves.
This is where Operational Excellence has an advantage. OE already gives organisations routines for reviewing performance, solving problems, improving processes, and sustaining gains. AI can strengthen those routines when it is introduced thoughtfully. It can help teams prepare faster, see patterns more clearly, and communicate more consistently. But the routine still matters. AI works best when it has a clear place in how the organisation manages and improves work.
Jobs are changing, so people need support
BCG’s 2026 research shows why this people-centred approach is becoming urgent. BCG estimates that over the next two to three years, 50 percent to 55 percent of US jobs will be reshaped by AI. That does not mean most jobs disappear. It means many roles will keep their title but change in how people perform the work, what tools they use, and what skills they need.
For operations leaders, this is a practical workforce issue. If jobs are being reshaped, people need support before the change becomes overwhelming. They need upskilling, clear expectations, and a realistic path from current ways of working to AI-enabled ways of working. Workforce strategy cannot sit at the end of AI transformation. It needs to be part of the plan from the beginning.
Measure value, not just tool usage
This also means organisations should measure adoption in a meaningful way. Counting logins or prompts may show activity, but it does not always show value. A better question is: are people using AI to improve work outcomes? Are they saving time? Are they improving quality? Are they communicating more clearly? Are they solving problems faster? Are they reducing avoidable admin? Are they making better decisions?
Good AI adoption measures should connect to operational value. For example, a team may track time saved in report preparation, reduction in manual rework, faster response to internal requests, improved quality of process documentation, or better completion of improvement actions. These measures keep the focus where it belongs: better work, not just more tool usage.
Trust comes from honest communication
Trust is another important part of adoption. People are more likely to use AI when they believe it is being introduced to help them do better work, not simply to monitor or replace them. Leaders should be honest about the purpose of AI. They should explain where it will support productivity, where it will change tasks, and where human judgement remains essential. Clear communication builds trust. Silence creates rumours.
Involve people in shaping AI use
The best approach is to involve employees in shaping AI use. Ask teams where they spend too much time on repetitive work. Ask which reports take too long to prepare. Ask where information is hard to find. Ask which tasks feel valuable and which tasks feel like unnecessary admin. These conversations help leaders find AI use cases that people actually care about.
This is very similar to good process improvement. People support change more strongly when they help shape it. They also know the work better than anyone else. They can tell leaders where AI might help, where it might create risk, and where a human touch is still needed. A people-centred AI approach respects that knowledge.
Operational Excellence gives AI a practical home
PATH OEMS™ fits naturally into this kind of thinking because it treats Operational Excellence as a managed system, not a collection of random improvement activities. If an organisation wants AI to support operations, it needs structure. It needs alignment, roles, routines, improvement discipline, and ways to hold the gains. AI can support the work, but the organisation still needs a system for managing the change.
Final takeaway: capable people turn AI into value
The main message is simple: prepare the people before expecting AI to create value. Give them permission. Give them guidance. Give them training. Give them practical use cases. Give them routines that make AI part of daily improvement. When people understand how to use AI confidently and responsibly, AI becomes more than a tool. It becomes a practical support for better operations.
Operational Excellence has always been about improving how work is done. AI does not change that purpose. It gives organisations a new way to support it. The organisations that benefit most will be the ones that help their people learn, adapt, and use AI with confidence. Better tools matter, but capable people turn those tools into better performance.
Leave a Reply