The Field That Saw This Coming: Why Knowledge Management’s Long Game is Now the Most Urgent Conversation in the C-Suite

Generation is now free. Curation is now everything. One field saw this coming.

The Inversion

For most of human history, production was the scarce thing. Creating content, performing deep analysis, or documenting a complex process required an enormous tax of effort, expertise, and time. We lived in a world organized around the bottleneck of making.

AI has completed the inversion of that world. The bottleneck hasn’t just widened; it has collapsed. What is scarce now is the opposite skill: the ability to know what to keep, what to sequence, what to frame, and what to let go. This is the latest evolution in the Digestion Gap lineage. If that earlier inquiry asked why organizations are drowning in more change than they can metabolize, curation is the survival mechanism. An organization that cannot curate cannot digest; it simply accumulates Cognitive Debt until the system stalls.

What Curation Actually Means

In popular discourse, “curation” has been diluted to mean picking a few interesting links for a social feed. Real curation is far more rigorous: a four-part practice of intellectual metabolism.

Selection is the first move, determining what is worth keeping at all. This requires a judgment of signal versus noise that is only possible with deep business context and a clear sense of purpose. Sequencing comes next: deciding the order in which knowledge is presented. The same facts arranged differently tell entirely different stories, and sequencing is the editorial intelligence that transforms a collection into an argument. Contextualization makes knowledge usable rather than merely findable, shaping it for a specific person in a specific situation. Knowledge without context is inert data. And finally, retirement: deciding what to release. This is the most neglected stage. Outdated knowledge is more dangerous than no knowledge because it actively misleads.

If this structure feels familiar, it should. This four-part practice is the knowledge lifecycle that the Knowledge Management field has been mapping for over three decades. You are arriving at a well-tended house.

The Field That Was Right Too Early

Knowledge Management emerged in the 1990s with a radical claim: organizational knowledge is a strategic asset, and managing it intentionally creates competitive advantage. For years, the field operated in a kind of principled obscurity, essential to the organizations that used it but largely invisible to the broader innovation conversation. KM practitioners were right. They were simply right too early.

That moment has passed. The frameworks KM developed are now the only things standing between enterprise AI and organizational incoherence. Consider the voices that have been leading this conversation for decades. Kate Pugh has documented what she calls “curation of exemplars,” the distinctly human work of discerning which knowledge is worth preserving and how to make it retrievable in the moment of need. Patrick Lambe uses knowledge auditing to map the invisible circulation systems of an organization, identifying where knowledge flows and where it gets stuck, an inquiry that connects directly to the “Energy Map” idea this newsletter has been developing. Nick Milton has spent his career demonstrating that capture without contextualization is just filing: his work on lessons-learned systems proves that curation must happen at the point of use, not just at the point of capture. And Heather Hedden provides what may be the least glamorous and most essential element, the taxonomies and ontologies that make curated knowledge findable. Curation without taxonomy is a library with no catalog.

AI has made these practitioners obsolete in no sense. It has made their decades-old arguments the most urgent conversation in the C-suite.

The AI Inflection Point

The current surge in AI capability has raised the stakes of curation, not resolved them.

Large language models and Retrieval-Augmented Generation (RAG) systems are only as reliable as the knowledge they draw from. When organizational repositories are uncurated, filled with outdated, uncontextualized, or contradictory content, AI systems produce confident, fluent, and wrong results. The problem is artificial confidence applied to uncurated material.

The KM field has been articulating the architecture this moment demands. Zach Wahl argues that enterprise AI performance is a direct reflection of knowledge quality and governance, framing Knowledge Management as the reliability layer that turns raw institutional data into something a system can be trusted to use. Andreas Blumauer’s work on knowledge graphs makes the same argument from the technical direction: the semantic layer beneath effective AI is curated knowledge, and without it, RAG systems draw from a chaotic inventory of unvetted documents rather than structured, governed truth. Maryam Alavi and Dorothy Leidner have formalized what is at risk: without deliberate KM governance, organizations face sensitivity leakage and embedded bias when legacy content is repurposed by AI without human oversight.

Curation, in this light, is critical infrastructure. It determines whether an organization’s AI investment returns strategic value or simply amplifies Cognitive Debt at scale.

What the Curator’s Edge Looks Like Across Roles

The Curator’s Edge is a concrete professional skill, and it shows up differently depending on where you sit.

For the individual knowledge worker, your curation practice is your expertise portfolio. What matters is what you have selected, organized, and can retrieve under pressure. AI can generate broadly and fluently, but it cannot judge relevance to your specific situation, your specific stakeholder, your specific moment. That judgment is the differentiating skill.

For the leader, curation of organizational memory, which stories get told, which decisions become precedent, which lessons are allowed to travel, is one of the least visible and most consequential things leaders do. Most leaders experience this as culture rather than curation. The distinction matters, because culture you can name, you can tend. Culture you cannot name tends itself, usually in directions you would not choose.

For the organization, the knowledge assets that AI depends on are either curated or they are noise. Organizations that delegated this to IT as a content management problem are discovering it was a strategic one. The Curator’s Edge, at organizational scale, is showing up in AI output quality. The organizations that invested in knowledge structure are watching their systems outperform those that did not.

The Translation Problem

KM practitioners have a precise vocabulary for all of this: tacit and explicit knowledge, communities of practice, knowledge transfer, knowledge audit. It is a vocabulary developed over decades of serious work. It has mostly stayed inside the discipline.

The broader professional world is arriving at the same ideas right now, naming them differently: institutional knowledge, AI grounding, context engineering, retrieval. Some of the translation is useful; new language reaches new people. Some of it is wasteful, reinventing concepts that already have names at the cost of the nuance the original terms carried.

The Curator’s Edge, as a professional posture, includes knowing which field has already done the thinking.

I’ve been framing curation as the next frontier of professional competence. KM practitioners would say it was always the frontier, and the rest of us just weren’t watching. For those in the field: what do you think has been lost in that gap? And what are you hoping this AI moment will finally make possible?

This continues the thread from The Digestion Gap: where that series asked why organizations can’t absorb what they consume, this essay asks who decides what is worth absorbing in the first place.

Madam I’m Adam