“The person who’s been at the company for 20 years knows that things aren’t written down anywhere. The AI doesn’t. This knowledge is not promptable. The interface between general AI capability and specific organizational reality is where value gets lost or captured.” – Nate B Jones – AI News & Strategy Daily
This observation captures a fundamental tension in the current wave of AI adoption: the gap between what large language models can do in theory and what they can accomplish within the messy, undocumented reality of actual organisations. Nate B. Jones identifies a critical vulnerability in the AI revolution-one that separates genuine competitive advantage from mere technological novelty.
The Problem of Tacit Knowledge
Jones points to a paradox that many organisations are only now beginning to recognise. A 20-year veteran of a company possesses something invaluable that no prompt can extract: tacit knowledge-the unwritten, often unconscious understanding of how things actually work. This includes informal decision-making processes, unspoken hierarchies, historical context that explains current practices, relationship dynamics, and the countless workarounds that keep operations running despite official procedures.
This knowledge exists in the gaps between what’s documented and what’s done. It lives in conversations, in muscle memory, in the collective understanding of teams. It’s the reason why onboarding a new employee takes months, not weeks, despite having an employee handbook. It’s why a seasoned manager can navigate a crisis that would paralyse someone following the official playbook.
Artificial intelligence systems, by contrast, operate on what can be tokenised and fed into a model. They excel at pattern recognition across documented information, at synthesising written knowledge, at optimising processes that have been explicitly defined. But they cannot intuit what isn’t written down. They cannot absorb the cultural knowledge that shapes decision-making. They cannot understand the political landscape that determines which ideas succeed and which fail, regardless of merit.
The Interface Problem
Jones frames this as an interface problem-the boundary between what AI can do and what organisations actually need. This is where the real value creation or destruction occurs. Consider a practical example: an AI system might be asked to optimise a workflow, but without understanding the informal approval process that actually determines whether a decision gets implemented, it will generate recommendations that look good on paper but fail in practice.
The interface problem manifests in several ways. First, there’s the documentation gap. Most organisations have far less documented than they believe. Policies exist on paper, but actual practice diverges significantly. An AI trained on official documentation will generate advice that contradicts how things are actually done. Second, there’s the context collapse. AI systems lack the historical understanding of why certain practices exist. A rule that seems arbitrary to an AI might exist because of a costly mistake made a decade ago that no one talks about anymore. Third, there’s the relationship blindness. Organisations are fundamentally social systems, and AI cannot perceive the trust networks, rivalries, and alliances that shape outcomes.
Nate B. Jones and the Evolution of AI Thinking
Jones has emerged as one of the most incisive commentators on the practical reality of AI deployment in knowledge work. His analysis distinguishes between AI capability-what the technology can theoretically do-and AI utility-what it can actually accomplish within real organisational constraints. This distinction has become increasingly important as organisations move beyond initial AI experimentation into genuine integration.
Jones’s broader framework emphasises what he calls high agency-the ability to act decisively despite uncertainty, to reframe obstacles as skill gaps rather than immovable barriers, and to use AI as a force multiplier rather than a replacement for human judgment. In the context of tacit knowledge, this means recognising that AI’s role is not to replace the 20-year veteran but to amplify their ability to codify and transmit what they know. The high-agency approach asks: “How can I use AI to bridge the gap between what I know implicitly and what the organisation needs explicitly?”
This perspective aligns with Jones’s broader work on second brains-AI-powered systems that don’t just store information passively but actively work to classify, route, summarise, and surface knowledge. The second brain concept, which Jones has evolved beyond earlier frameworks like Tiago Forte’s CODE methodology, recognises that the future of knowledge work lies not in replacing human expertise but in creating systems where human insight and AI capability work in concert.
The Broader Context: Tacit Knowledge in Organisational Theory
The challenge Jones identifies has deep roots in organisational theory and knowledge management. The distinction between explicit knowledge (what can be documented) and tacit knowledge (what is embodied in people and practice) was formalised by Michael Polanyi in the 1960s with his famous observation: “We know more than we can tell.” This insight became foundational to understanding why knowledge transfer is so difficult and why organisations lose critical capabilities when experienced people leave.
Ikujiro Nonaka and Hirotaka Takeuchi built on this framework in their theory of organisational knowledge creation, arguing that the most valuable knowledge emerges through the interaction between tacit and explicit forms. They identified four modes of knowledge conversion: socialisation (tacit to tacit), externalisation (tacit to explicit), combination (explicit to explicit), and internalisation (explicit to tacit). The challenge for organisations deploying AI is that most AI systems operate primarily in the combination mode-they’re excellent at working with explicit knowledge but cannot participate in socialisation or externalisation without human intermediaries.
This is where Jones’s insight becomes strategically important. The organisations that will capture value from AI are not those that attempt to replace human knowledge with AI systems, but those that use AI to make tacit knowledge more accessible and actionable. This requires intentional effort to externalise what experienced people know, to document the undocumented, and to create systems where AI can help surface and apply that knowledge at scale.
The Knowledge Capture Challenge
The practical implication is that knowledge capture becomes a competitive advantage. Organisations that can systematically convert the tacit knowledge of their most experienced people into forms that AI can work with-whether through documentation, structured interviews, decision frameworks, or process mapping-will be able to scale that expertise. Those that cannot will find their AI investments generating plausible-sounding but contextually inappropriate recommendations.
This is not a technical problem alone. It’s an organisational and cultural challenge. It requires creating space for experienced people to articulate what they know, building systems that reward knowledge sharing rather than hoarding, and recognising that the 20-year veteran’s value increases rather than decreases in an AI-augmented environment. Their tacit knowledge becomes the training data for organisational intelligence.
Jones’s framing also highlights why generic AI solutions often disappoint. A general-purpose AI system, no matter how capable, cannot understand your organisation’s specific reality without significant human interpretation and guidance. The interface between general capability and specific context is where organisations must invest their effort. This is where the real work of AI adoption happens-not in implementing the technology, but in bridging the gap between what AI can do and what your organisation actually needs.
Implications for Knowledge Work
For knowledge workers and leaders, this insight suggests a reorientation of priorities. Rather than asking “How can AI replace this function?”, the more productive question is “How can I use AI to make my tacit knowledge more valuable and more widely applicable?” This aligns with Jones’s broader emphasis on agency-the ability to shape how technology serves your goals rather than being shaped by it.
The organisations that thrive in the next phase of AI adoption will be those that recognise tacit knowledge not as an obstacle to automation but as a strategic asset to be systematically developed, documented, and amplified through AI systems. The 20-year veteran doesn’t become obsolete; they become essential to the process of making AI genuinely useful.
References
1. https://globaladvisors.biz/2026/01/30/quote-nate-b-jones-on-second-brains/
3. https://www.natebjones.com/prompts-and-guides/products/second-brain
4. https://www.youtube.com/watch?v=Td_q0sHm6HU
