Tacit Space Shrinkage: How AI Makes More Expertise Explicit

As AI improves, more knowledge can be expressed, operationalized, and reused. That changes how organizations think about expertise and governance.

Every organization has knowledge that nobody can quite put into words. The machinist who “just knows” when a tolerance is off. The sales director who senses when a deal is going sideways. The nurse who recognizes deterioration before the monitors do. We call this tacit knowledge, and for decades we’ve treated it as a fixed feature of organizational life. Valuable, irreplaceable, and resistant to capture.

As AI capabilities expand, less knowledge remains practically inexpressible. Much of what we called tacit knowledge was blocked by the knower’s representational limits, although some knowledge resists representation itself.

The distinction changes what organizations can expect from expertise, training, and governance.

This argument builds on two foundational claims. First, knowledge is capability to produce outcomes. That’s an operational definition that sidesteps epistemological questions about belief and justification. Second, tacitness is agent-relative. It’s a property of transfer relationships between agents, not an intrinsic feature of knowledge itself. Together, these foundations make “tacit domain shrinkage” coherent: as AI capabilities expand, transfer paths open that didn’t exist before.

Nonaka’s SECI model describes externalization as converting tacit knowledge into explicit knowledge, changing its state through human-to-human transfer. Agent-relative tacitness instead describes which agents can access a capability. AI introduces different perceptual and representational abilities, allowing knowledge tacit to humans to become accessible to a system without becoming explicit in the SECI sense. “Tacit space shrinkage” names that expansion in the agents able to use the knowledge.

Two Kinds of Tacitness

Michael Polanyi described tacit knowledge as knowing more than we can tell. A cyclist cannot fully articulate how she balances; a chess master cannot exhaustively explain his positional intuition.

Organizational tacit knowledge includes two distinct phenomena.

Structurally tacit knowledge depends on integrated bodily skills, perceptions, and value commitments that make explicit statements meaningful. Trying to articulate every element can fail or degrade performance by disrupting that integration. This is Polanyi’s tacit dimension.

Representationally tacit knowledge can be expressed without losing its usefulness, but the knower lacks the vocabulary or concepts. A process engineer may recognize drift in a system without knowing the control-theory terms for it. Appropriate concepts or an interlocutor can help her articulate that recognition.

Much organizational knowledge is representationally tacit: people have yet to express it because they lack the words, although the knowledge could be expressed.

LLMs as Expressive Prosthetics

Large language models can help experts express knowledge they have previously been unable to articulate.

In that work, an LLM can:

• The model proposes candidate formulations when the expert “knows what they mean but cannot say it”
• It offers terminology and distinctions from other domains that might capture what the expert is trying to express
• It suggests ways to structure process descriptions or narratives
• It generates examples and counterexamples that help sharpen explanations

This directly addresses the expressive-deficit problem. Knowledge that felt ineffable to an individual knower becomes articulate once the model supplies concepts and language they didn’t know were available.

The test: If an expert, working with an interlocutor with broader vocabulary and conceptual repertoire, can iteratively articulate knowledge until a competent third party can act on it, the knowledge was likely representationally tacit. If articulation fails even with extensive assistance, if “you just have to do it yourself to know,” the knowledge may be structurally tacit.

LLMs can propose successive formulations using vocabulary and conceptual frameworks from across human knowledge. Their range and capacity for repeated attempts make this test practical in ways it was not before.

Learning from Showing, Not Just Telling

Multimodal models can also acquire capabilities from demonstrations.

Many classic examples of tacit knowledge involve “you have to see it done” or “you learn it at the elbow of a master.” The knowledge was tacit in part because the only transmission mechanism was experiential. Watching, imitating, practicing under guidance.

Multimodal models ingest video, audio, and sensor streams. They can process vastly more demonstrations than any human apprentice, identify patterns across them, and encode those patterns into representations that can be deployed, transferred, and refined.

A master’s demonstration becomes training data. Models can encode knowledge that humans could transfer only through experience, making it explicit for the system.

The Governance Challenge

As knowledge moves from tacit-for-humans to explicit-for-systems, it doesn’t necessarily become transparent-for-humans. A model can encode the expert’s pattern recognition into weights and parameters that produce reliable outputs. But those weights may not be interpretable. The knowledge is effectively explicit (reproducible, transferable, operationalizable) without being inspectably explicit (readable, auditable, contestable by humans).

This creates a new governance gap. Traditional tacit knowledge was at least located in identifiable humans who could be consulted, questioned, and held accountable. System-level knowledge may be operational without being legible. The organization can reliably act on it without anyone being able to fully explain it.

We routinely operate complex systems without fully understanding them. Governing these systems still requires assigning accountability, auditing processes we cannot fully inspect, and providing ways to contest decisions based on patterns humans cannot consciously identify.

The Structurally Tacit Human Core

The practically ineffable region for organizations is shrinking. Knowledge that was blocked by expressive deficits is yielding to LLM-assisted articulation. Knowledge that was transmitted only through demonstration is yielding to multimodal learning. Knowledge that was locked in individual expertise is yielding to systems that can detect, encode, and deploy patterns at scale.

A structurally tacit human core remains: lived experience, value commitments, the felt sense of what matters, and the integration of perception, emotion, and action in skilled judgment. Better representational tools cannot resolve these.

More of what organizations have called “tacit” than they assume was representationally blocked, not structurally inexpressible. Those blocks are dissolving. The boundaries of the articulable are moving faster than most leaders recognize, and the organizations that grasp this will capture knowledge their competitors still believe is uncapturable.

Some of the machinist’s feel for tolerance, the sales director’s deal intuition, and the nurse’s pattern recognition is irreducibly human. Much can be articulated with the right assistance. Answers to Critics explains the limits of this claim.