An Organization Doesn’t Become AI-Native Through a Course — but Through the People Who Join It

I have seen organizations run three separate AI courses for the entire development team, invest in training days, bring in consultants, and run pilots — and end up in exactly the place where they started. And I have seen organizations that added to the team a single developer who already worked in an AI-native way from day one, and within weeks the way the whole team worked began to change. The difference between these two scenarios is one of the most important lessons about AI transformation: ways of working do not travel between slide decks, they travel between people.

This is why so many organizations, despite a sincere investment, get stuck in an AI-Ready state and fail to become AI-Native. They treat the change as a knowledge project that can be delivered through training, when in practice it is a change in a way of working — and that, by its nature, travels differently.

Why Training, Consultants, and Pilots Are Not Enough

The data illustrates the gap. On one hand, the adoption of AI tools has become almost universal, and most organizations have already given employees basic knowledge about using them. On the other hand, a study published by MIT’s Project NANDA found that only about 5% of the organizations examined reached measurable business impact from AI, measured over six months. The figure deserves to be treated with care — the definition of success was narrow and the measurement window short — but the direction is clear: the knowledge arrived, the way of working did not.

The reason for this runs deep. Training transfers information: which tools exist, which commands to type, what capabilities there are. But a way of working is not information. It is a collection of habits, small decisions, unwritten norms, and gut feelings about when to trust the tool and when not to. These cannot be conveyed in a slide deck, just as you cannot learn to ride a bicycle from a manual. An isolated pilot suffers from the same problem: it creates an island of new activity that fails to infect the rest of the organization, because there is no one to carry the way of working onward.

Why Isolated Pilots Fail to Infect the Organization

The pilot is perhaps the favorite tool of organizations trying to become AI-native, and it is also one of the most misleading. The logic sounds right: we will take a small team, let it work in a new way, prove value, and then scale to the whole organization. But in practice, many pilots succeed within themselves and fail at exactly the scaling stage. They produce an island of new activity that fails to spread to the continent around it.

The reason goes back to the same principle: a way of working does not travel through a success report. Even when the pilot proves excellent results, the results themselves do not carry with them the way of working that produced them. The rest of the organization sees the numbers but not the habits, decisions, and norms behind them. Without people who carry the way of working onward — physically, in shared work — the pilot remains an isolated pocket that fades the moment managerial attention moves on.

This is why organizations that rely on pilots as a transformation strategy find themselves repeating the same cycle again and again: a successful pilot, a scaling that stalls, and a new pilot. That cycle can go on for years without the organization as a whole moving from Ready to Native, simply because the chosen mechanism does not fit the nature of the change.

Ways of Working Travel Between People, Not Between Slide Decks

To understand how real change happens, it helps to think about the idea of knowledge grafting. Unlike the transfer of information, in which someone explains to someone else what to do, knowledge grafting happens when a person who already works in a certain way joins a group and begins to work beside it. The group absorbs the way of working not through explanation but through daily exposure: they see how he plans, how he checks AI output, when he stops, and how he talks about those decisions.

This is the same mechanism by which professional knowledge has always been passed on — apprenticeship, not lecture. The new way of working seeps through countless small interactions, until it becomes the team’s norm without anyone having defined it formally. It is slower than a few hours of training, but it is the only thing that truly changes the way the work is done.

This distinction between transferring information and grafting knowledge also explains why so much investment in training yields so little change. Information can be documented, replicated, and distributed quickly — and that is exactly why it creates no advantage: if it can be taught in a course, competitors can learn it just as easily. A way of working, by contrast, is hard to transfer and therefore hard to copy, and precisely for that reason it is a source of sustainable competitive advantage. An organization that has built an AI-native way of working through people holds an asset competitors cannot simply download or buy — they have to build it themselves, person by person.

Knowledge Grafting: How One AI-Native Developer Changes a Whole Team

When a developer who works in an AI-native way joins a team, the change begins at the margins and spreads inward. Code review is usually the first place: he brings with him different expectations about what to check and what not to, and raises questions the team was not used to asking. Planning changes when he shows that a prototype can be produced in hours and a decision made on its basis, rather than discussing assumptions for days. Norms change when he demonstrates, again and again, when he chooses not to use AI and why.

The interesting thing is that all of this happens without anyone defining it as a transformation project. There is no document, no change plan, no steering committee. There is simply a person who works differently, and a team that learns from him in the course of shared work. This is knowledge grafting in action, and it is effective precisely because it does not feel like an organizational initiative but like ordinary work.

Brain and Muscle: A Local Core and Global Teams

This insight takes on strategic meaning when you think about the structure of the team as a whole. In a model taking shape at many organizations, the local core — often in Israel — serves as the “brain”: it holds the business context, the institutional knowledge, and the judgment about where AI stops and which decisions demand a deep understanding of the product and the customer. This is a role that cannot and should not be outsourced.

Alongside the core, global teams that already work in an AI-native way serve as the “muscle”: they bring with them not only execution capacity but the way of working itself. When such a team integrates alongside the local core, it channels AI-native practices inward through shared daily work — the very same knowledge grafting, but at the scale of a team rather than an individual. The combination of a local brain and a global muscle lets the organization cross the gap from Ready to Native faster and with lower risk than any course or pilot.

It is important to stress that this is not a move to replace the core but to strengthen it. The local core remains the center of gravity for decisions; the global teams expand capacity and bring ways of working, but always within the context and judgment the core defines. It is a delicate balance that requires a culture of trust, transparency, and clear accountability — but when it works, it turns the very act of building the team into the engine of transformation.

The Next Step Is Not Another License — It Is the Next Person

If an organization finds itself stuck at AI-Ready despite widespread adoption of tools, the answer is almost never another license, another tool, or another course. These add more information to an organization already saturated with information. The answer usually lies in people: who joins the team, which way of working they bring, and how many opportunities the existing team has to learn from them in the course of work.

This is why, in the AI-native era, the workforce strategy becomes the AI strategy. The decision about who will join the next team is not just a hiring decision — it is a decision about the way of working the organization will adopt. Organizations that understand this stop looking for the next tool and start looking for the next person who will bring with them the way of working they want to embed.

The Mistake of the “Transformation Project”

One common mistake is trying to turn the move to AI-native into a managed project: a steering committee, milestones, adoption targets, and a quarterly status deck. That approach fits classic technology projects, but it misses the nature of the change. A way of working is not a deliverable you can “ship” by a due date; it is a culture that emerges through practice. The more the organization manages the change as a project, the further it drifts from its essence.

This does not mean there is no place for intent and resources. On the contrary — you need to invest, but in the right places: in hiring the right people, in opportunities for shared work, and in legitimizing a change of norms from the bottom up. The investment shifts from “how will we teach everyone” to “whom will we bring in, and how will we let their way of working spread.” It is a subtle shift in perspective, but it separates an organization that gets stuck from one that moves forward.

Conclusion: A Transformation That Travels Through People

The move from AI-Ready to AI-Native is not bought with training and not achieved through pilots. It happens through people who bring a way of working with them, and through teams that learn from them in the course of doing. Knowledge grafting — the transfer of practices through daily exposure rather than through slide decks — is the mechanism that truly changes the way the work is done. Organizations that keep looking for the solution in tools will stay stuck. Those that understand that transformation travels through people, and build their team accordingly, are the ones that will truly become AI-Native.

Frequently Asked Questions

Why isn’t AI training enough to make an organization AI-Native?

Because training transfers information — which tools exist and how to use them — while a way of working is a collection of habits, decisions, and unwritten norms. These cannot be conveyed in a slide deck, just as you cannot learn to ride a bicycle from a manual. That is why organizations with high adoption and low return discover that the knowledge arrived but the way of working did not change.

What is knowledge grafting?

It is a mechanism in which a person who already works in a certain way joins a group, and the group absorbs the way of working through daily exposure rather than through explanation. They see how he plans, checks AI output, and decides when to stop — and learn from it in the course of work. It is the same apprenticeship mechanism by which professional knowledge has always been passed on.

How does a global team accelerate the move to AI-native?

A global team that already works in an AI-native way brings with it not only execution capacity but the way of working itself. When it integrates alongside a strong local core that holds the business context, it channels AI-native practices inward at the scale of an entire team — the same knowledge grafting, but broader and faster.

How many AI-native developers does it take to change a whole team?

Sometimes one or two are enough, provided they work closely with the team and not in isolation. The decisive factor is not the quantity but the degree of exposure: how many opportunities the rest of the team has to see the way of working in action and learn from it. A whole global team accelerates the process, but a focused presence can also ignite change.

About Global Teams

Global Teams focuses on linking top-tier technical professionals to forward-thinking organizations worldwide. With our unique recruitment approach, frameworks for cultural integration, and management strategies aimed at retention, we help companies assemble high-performing, AI-native technical teams while achieving 40–60% cost savings. Arrange a consultation with our technical talent strategists for a tailored talent acquisition plan.