KM’s Role in Digital Transformation

Digital transformation promises much but often delivers less than expected. It seems almost silly in the era of AI to discuss digital transformation anymore, as most organizations have likely already undergone digital transformation. They should now focus on retransformation.

Of course, we have been through a “re” before, during the reengineering revolution. I wrote then that it was impossible to reengineer something that wasn’t engineered in the first place. We have moved beyond that. Organizations have become overwhelmingly digital, but most have not become effectively digital.

Organizations have become overwhelmingly digital, but most have not become effectively digital.

Organizations treat technology as if it alone will realign work, shift company policies and practices, and create value. That organizations are still talking about digital transformation is evidence that it has become just a lifecycle metaphor for tool acquisition layered on top of old habits. The rhetoric still suggests inevitability, but effective use of technology requires discernment and judgment.

Knowledge Management (KM) should sit at the center of this tension. KM has always insisted that people, processes, and practices anchor transformations, and that perspective remains critical. Technology, the new and shiny, the easily acquired, often outweighs those considerations in transformation programs. KM’s practices for reflection and collaboration can help organizations discover how to become effective at using technology, even if they aim to accomplish great things at great speed.

KM becomes a steadying force, perhaps seen by some as a throttle, but I see it more as a moment to seek reason when speed leans toward the edge of chaos, and uncertainty becomes context.

The need for rational reflection has not shifted simply because AI and machine learning are now part of the mix. If anything, the arrival of generative AI reinforces KM’s original point: without governance, context, and trust, AI only magnifies noise. The promise of transformation without policy and practice alignment or thoughtful integration is the reason even those who have digitally transformed continue to discuss digital transformation rather than strategic alignment, execution and navigation.

Effectively digital means leveraging technology as a tool to execute strategy, not as a topic that requires a strategy.

AI introduces new dynamics to our relationship with information. Intelligent search, summarization, and recommendation systems can accelerate access to knowledge, yet they are built on fragile foundations. Without robust taxonomies, metadata, and curation, AI models are left to infer meaning from unstructured sources. They may generate plausible answers that dissolve under scrutiny.

Knowledge graphs illustrate both potential and risk. When well-designed, they create a semantic web that links structured and unstructured information. When poorly managed, they become brittle, encoding outdated assumptions that AI will repeat without question. KM must treat these tools not as replacements for human sense-making, but as amplifiers that require active stewardship.

Remote and hybrid work have made collaboration platforms essential; however, the temptation is to view them as ends in themselves. KM’s role is to ensure these environments support authentic communities of practice, enable expertise location, and capture tacit insights.

AI can enhance collaboration by surfacing expertise and recommending connections, yet it can also distort organizational dynamics if treated as an oracle of truth. A recommendation engine may privilege the already visible and marginalize the less documented but equally valuable knowledge.

At the same time, AI introduces new collaborators, and I say that as a plural, because the effective use of AI may not be a single system for analysis or insight, but leveraging multiple systems as adversaries to challenge and explore the edges of ideas that may fall out by humans working in stressful work environments.

The deeper question is how much organizations want to delegate social connection to algorithms. The answer should be pretty clear. While AI can mirror the codified, it can’t gain access to what people haven’t shared, or are just in the process of learning, without people to facilitate the transformation of data, experience and context into knowledge. KM recognizes that meaningful collaboration depends on trust, relationships, and human choice; and only meaningful collaboration leads to collective knowledge.

The best test of KM lies in process integration. Knowledge practices must become embedded in daily workflows, rather than being appended as afterthoughts. AI assistants that automate routine knowledge tasks—such as drafting reports, tagging documents, and summarizing conversations—can add significant value. But they also introduce opacity. If AI-generated insights influence decisions, organizations must demand transparency, provenance, and the ability to interrogate the logic behind recommendations.

This skepticism should not be read as dismissal. The efficiency gains are real, but without accountability, automation risks eroding trust rather than building it. KM must frame AI not just as a convenience, but as part of a governed knowledge ecosystem.

Digital transformation is often presented as a destination. In practice, it is a continuous negotiation between technological affordances and organizational realities. Change fatigue is real, and workers respond best when they see direct, personal benefit. KM practitioners are experienced in articulating this benefit because they have long wrestled with the “what’s in it for me?” question.

The language of transformation implies finality, turning one thing into something else, but because technology continues to evolve, transformations likewise need to remain continuous.

There is no future proofing, and technological maturity is a myth. Maturity models and end-states suggest stability that rarely exists in practice. A more honest approach is to recognize digital change as a series of experiments, adjustments, and accommodations —a pattern of adaptation rather than a linear march.

And keep in mind that the arrival of new technology doesn’t cause the need for digital transformation; instead, it creates new expectations, informed and driven by social, economic, environmental, and political factors.

Knowledge remains a durable asset. Technology, whether enterprise resource planning, resume scanners, collaboration platforms, or AI agents, should be judged by how well it preserves, shares, and extends that knowledge.

The task is not to replace human intelligence with machines but to design systems that support collective sense-making. AI is best seen as a partner in that effort, one that demands human oversight to remain useful.

Digital transformation proves less about radical reinvention than about creating conditions for knowledge, individuals, social networks, and practices to thrive amid shifting circumstances.

Organizations that succeed will be those that accept the limits of tools, invest in policy and practice, and treat knowledge as both a process and an asset, and recognize that they aren’t seeking to be transformed, but rather that they are living in a constant state of transformation.