The AI didn’t hallucinate. It tripped on a path nobody laid properly.
Why ‘the AI hallucinated’ is almost always the wrong diagnosis, and what’s actually breaking underneath.
Knowledge Management
7 posts
Your KMS has a taxonomy. Your enterprise does not.
Why the semantic layer I spent years dismissing turns out to be the thing missing everywhere else.
You don't have a knowledge base. You have a filing cabinet.
You pointed your AI at SharePoint and now it's confidently wrong. Why? Because you don't have a knowledge base. You have a filing cabinet. And until you understand the difference between a document repository and a KMS, your AI project is retrieving the wrong things, faster, at scale, with confidence.
The knowledge base is auditable. The AI isn't.
A legal dispute, a standard fifteen-minute evidence request, and the moment it stopped being standard.
Data Permissions vs Knowledge Governance
The risk isn't that enterprise AI search tools don't work. It's that they work exactly as designed, for knowledge workers, not contact centres. When you use discovery infrastructure to do governance work, the permissions get respected. The governance does not.
The Unsexy Truth About AI and Knowledge Management
Why the invisible infrastructure nobody wants to build is still the foundation of success.
The AI Paradox in Knowledge Management: When Innovation Meets Fear
Organisations are excited about AI knowledge delivery. And terrified to actually use it. The paradox isn't a technology problem; it's a change management problem.