AI adoption has moved fast. Access is no longer the constraint it once was: employees across most organizations can now reach for powerful AI tools whenever a task calls for them. That shift happened quickly, and it is worth pausing on, because it changes what the next phase of AI work actually requires.
Access was never the hard part. Consistency is. The same tool, used by two different people on two similar problems, can return excellent output in one case and something generic in the other. That difference shows up constantly, and it is easy to mistake it for a tooling problem or a training problem. It is usually neither.
At The Kendall Project, we think the real explanation is more specific: AI models are extraordinarily capable, but they do not arrive already knowing an organization's customers, processes, policies, systems, or the exceptions that make real work real work. Every employee ends up supplying pieces of that picture on their own, in their own words, with their own gaps. The result is a widening space between what AI can theoretically do and what it can reliably do inside a specific business or problem area. We call that space the Context Gap.
Closing it is not a matter of better prompting tips or a longer pilot list. It is a different kind of work, and it is a Team Sport: capturing what an organization already knows, structuring it, validating it, and keeping it current. We call this discipline Context Operations, and we think it is the next real competitive layer in enterprise AI, one that sits above the model and the license, and determines whether AI will deliver real, consistent value for your organization.
This is why "muscle" is the right word for what comes next, rather than "tools" or "adoption." Tools can be purchased in an afternoon. Muscle is built through repetition, and it compounds: the more context an organization captures and validates, the less frequently a new project has to start from zero. Skipping that work does not make the AI tools less capable, it just makes AI less impactful for your organization. It makes AI's capability harder to depend on, project by project, team by team. And because models, platforms, and interfaces will keep changing at a breakneck speed, an organization's own context is the one asset that carries forward through every future upgrade. We call that Context Modularity and that is the part worth building deliberately now, rather than assuming it will accumulate on its own.
In our new whitepaper, Beyond Experimentation: It's Time to Put On AI Muscle, we lay out:
- Why the current adoption gap is a context problem, not a technology problem or a training problem
- What Context Operations actually involves, and what it deliberately does not mean (hint: it is not "upload everything and hope")
- A three-level framework for building AI capability, from the individual employee up through the team and the company
- A sequenced set of steps leaders can start on now, without waiting for a full transformation plan
The next two years will not be won by whoever experiments the most. They will be won by the organizations that turn what they have already learned into something repeatable.