Since the 1990s, the management of intellectual capital in companies has been a theoretically sound and well-researched concept. Japanese researchers Ikujirō Nonaka and Hirotaka Takeuchi found that the constant transformation of these resources is crucial. This is the only way to create long-term added value, which is based on implicit insights and intuition, for example. Accordingly, four different phases are decisive: socialization, externalization, combination and internalization.
Socialization transfers tacit knowledge between individuals through shared experience. Think of an apprentice learning from a master craftsman, or a new executive absorbing company culture through daily interactions. Externalization converts tacit knowledge into explicit, documented form. Combination reorganizes and integrates explicit knowledge into broader systems. Internalization turns explicit knowledge back into tacit knowledge through practice and experience. Most organizations handle combination and internalization reasonably well. The bottleneck has always been externalization: capturing the unspoken expertise that experienced employees carry but cannot easily articulate.
Technology plays a key role in this process, and artificial intelligence is no exception. Thanks to new approaches such as multimodal language models, even the externalization of implicit knowledge can now be technically implemented. These models can analyze how experts interact with complex documents, observe decision patterns across thousands of cases, and codify the resulting insights into structured knowledge that the entire organization can access. Our CEO Christopher Helm describes exactly how this works in a German-language Guest article for the trade magazine Digital Business Magazine.
The most important insights
- Until now, implicit knowledge was difficult to use technically. It existed only in the minds of experienced employees and was lost when they left the organization
- However, AI can accelerate knowledge transformation across the board, reducing the time from knowledge capture to organizational availability from months to days
- The multimodal voice models used for this require a high level of connectivity and precise integration with existing enterprise systems and data sources
- This means that even highly complex information carriers such as reports or presentations can be used as inputs for knowledge extraction and structuring
- The development expertise of AI providers is particularly important for this, as off-the-shelf solutions rarely handle the domain-specific nuances that make tacit knowledge valuable
About Digital Business Magazine
Digital Business Magazine is a German-language trade magazine that covers topics relating to digital transformation and automation in companies. It is aimed at entrepreneurs, board members, IT and HR managers. The publication focuses on practical applications of emerging technologies in enterprise environments, with particular emphasis on how German and European companies can leverage digital tools to improve operational performance and competitive positioning. Guest contributions from practitioners like our CEO reflect the magazine's commitment to real-world expertise over theoretical discussion.