Technology to connect people and knowledge

In the evolving world of Knowledge Management, technologies that unify distributed knowledge—such as generative AI, knowledge graphs, and vector databases—are revolutionizing how organizations empower their workforce. As Jelani Harper writes in KMWorld,

“The technologies and approaches shaping the future of knowledge management are those that enable enterprises to make information-based decisions more quickly, reliably, and transparently.”
(KMWorld)

This vision is coming to life through innovations like natural language interfaces, retrieval-augmented generation (RAG), and metadata-aware knowledge graphs.

“Democratizing” Knowledge for All

Jans Aasman, CEO of Franz, captures the essence of today’s democratized interfaces:

“‘Democratize’ is exactly the right word in this case, because anyone, without having to learn a complex language or have a computer science degree, can now talk to a knowledge graph.”

This shows how modern KM tools make powerful insights accessible to all employees—no coding required.

Trustworthy Summarization: Powerful but Cautious

While language models are increasingly used for document summarization, Aasman emphasizes the need for attention to accuracy:

“You can never trust it.”

He also notes when such tools are best applied:

“People say if the price of making a mistake is not high, and you’re doing something that’s very repetitive, like reading 1,000 restaurant reviews, … that’s the best way to use an LLM.”

Bridging Domain Silos with Natural Language

Aasman illustrates how language models can seamlessly interpret metadata across departments:

“People from different domains, that don’t know what the data in the other group looks like, can still ask questions, and the LLM will understand what you’re trying to ask and translate it into something that can be queried from the database.”

This breakthrough enables legal, engineering, and other teams to communicate intuitively across structured knowledge systems.

Enhancing RAG with Metadata

As we increasingly rely on retrieval-augmented generation (RAG), Aasman highlights a common pitfall:

“If you only do RAG, I’ve found that just popping 100,000 texts into a vector store gives bad results because you miss all the metadata per element.”

Integrating rich metadata—or full knowledge graphs—ensures more accurate, contextually relevant outputs from AI systems.

Why These Insights Matter

Lower barriers to knowledge access: By enabling natural language queries, organizations empower broader staff engagement with complex systems.

Balanced innovation: Aasman’s candid view on trust in LLMs helps set realistic expectations and safe use cases.

Cross-functional clarity: Bridging lexical gaps across domains fosters smoother collaboration and decision-making.

Richer AI outcomes: Metadata-enriched RAG delivers smarter, more reliable information than metadata-free setups.

Looking Ahead: KM in the Age of AI

Jans Aasman’s perspectives illuminate a future where KM tools must balance accessibility, accuracy, and context. From democratizing access to intelligently integrating metadata—these principles guide us toward knowledge systems that are both powerful and trustworthy.