Published: October 7, 2026 by Dr. Norman Mooradian

As artificial intelligence becomes increasingly embedded in organizations, managers, information professionals, and knowledge managers face an important question: should AI be used primarily to replace human knowledge work, or should it be used to support and augment human knowledge?

This question is at the center of my recently published article, “KM Knowledge Models and Epistemologies: Implications for AI Integration Strategies,” in Philosophy of Management. In the article, I examine how knowledge management (KM) can help us think more carefully about AI integration, especially when organizations are deciding whether to pursue substitution-based automation or human-centered augmentation.

Much of the current conversation about AI focuses on efficiency, productivity, and automation. These are important considerations, but they do not fully capture what is at stake when AI is introduced into knowledge-intensive environments. Knowledge work is not simply the processing of information. It often involves judgment, interpretation, experience, tacit understanding, creativity, and the ability to act responsibly within complex organizational and social contexts.

This is where knowledge management offers a valuable lens. KM has long emphasized that knowledge is not just a technical asset to be captured and stored. It is also deeply human, social, contextual, and practical. KM theories and models have incorporated epistemological questions—questions about what knowledge is, how it is created, how it is justified, and how it is shared. These questions are especially relevant in the age of AI.

In the article, I explore several foundational KM models and epistemic concepts, including the DIKW hierarchy, the SECI model of knowledge creation, and theories of tacit knowledge. Each of these models helps illuminate different aspects of the relationship between human knowledge and AI systems.

The DIKW hierarchy—data, information, knowledge, and wisdom—helps distinguish between different levels of cognitive and organizational value. AI systems are often very powerful at processing data and generating information. However, moving from information to knowledge, and especially from knowledge to wisdom, involves interpretation, judgment, and context. This distinction is important because organizations may overestimate what AI systems can do if they treat all knowledge work as reducible to information processing.

The SECI model, developed by Nonaka and Takeuchi, highlights the dynamic process through which tacit and explicit knowledge interact. Knowledge creation involves socialization, externalization, combination, and internalization. This model reminds us that knowledge often emerges through interaction, shared experience, and organizational context. AI may support parts of this process, but it does not eliminate the need for human participation in knowledge creation.

Tacit knowledge is another major theme of the article. Much of what people know is difficult to fully articulate or codify. It may be embodied in practice, developed through experience, or shaped by participation in a professional or organizational community. This kind of knowledge presents a challenge for replacement-oriented AI strategies. If an organization assumes that knowledge can be completely extracted, encoded, and automated, it risks misunderstanding the nature of the work itself.

The article argues that KM generally supports a human-enabling or augmenting approach to AI integration. This does not mean that automation has no place. Some routine or well-structured tasks may be appropriate for automation. However, when work requires tacit knowledge, contextual judgment, abductive reasoning, ethical discernment, or practical wisdom, AI should be designed and implemented in ways that enhance human capabilities rather than replace them.

This has important implications for information professionals. Librarians, archivists, records managers, data professionals, and knowledge managers all work in environments where AI tools are becoming more prevalent. These professionals are well positioned to ask critical questions about how AI systems represent knowledge, how they support or undermine human expertise, and how they affect accountability, trust, and organizational learning.

For iSchool students and faculty, the article also points to the importance of connecting technical questions about AI with philosophical and ethical questions about knowledge. AI integration is not only a matter of choosing tools. It is also a matter of deciding what kinds of knowledge organizations value, what kinds of human expertise they want to preserve, and what kinds of futures they want to create for knowledge work.

Ultimately, my goal in this article is to show that knowledge management provides both theoretical and practical resources for navigating AI adoption. KM encourages us to move beyond a narrow view of automation and toward a more thoughtful understanding of how humans and AI systems can work together. As organizations continue to experiment with AI, the field of KM can help ensure that these efforts remain grounded in a rich, human-centered understanding of knowledge.