Knowledge & Decisions

Organisations make decisions with the knowledge that reaches the room, and much of that knowledge sits with people who are not in it. AI is now changing how that knowledge moves, collecting it and handing it to whoever asks.

The difficulty is that sharing knowledge is a choice. People can share when they feel recognised and hold back when they feel used, and an AI system that takes their answers without credit can quietly change that choice.

My PhD, completed at Manchester Metropolitan University, studied knowledge-sharing behaviour: how people perceive sharing, what they intend to do, what they actually do, and what enables or restricts it. It drew on the Theory of Planned Behaviour, which links intentions to actions, and the SECI model of how knowledge moves between tacit and explicit forms. The setting was digital entrepreneurs running small and medium-sized businesses in the UK and Germany, in low-code and no-code communities. In that sample, constructive feedback and recognition of expertise encouraged participation, while low-quality or self-promotional contributions could restrict it. The findings cover that sample only.

The question I carry into AI work starts there. AI models are trained on what people chose to share, and they use and exchange it at a scale no community ever did. So when AI becomes part of a workflow, what happens to what people know, whom they trust and the decisions they make? The thesis does not answer it. It supplies a lens, and each application needs new interviews or a labelled scenario.

On the practical side, Decide Better offers fifty compact frameworks for framing, testing and writing down a choice before it is made.