Writing

Articles and reflections.

Notes from building AI platforms in a regulated industry: architecture, evaluation, governance, and the people who make transformation work. RSS

Measuring AI: what to count, and how

AI transforms each organisation differently, so its value has to be measured, not assumed. Four indicators for every use case, one guardrail to protect what makes the business different, and how to measure both honestly.

measurementstrategygovernanceleadershipgenai
article

Staying in control when AI writes the code

SaaS tools, AI providers and agents now change faster than people can absorb. What control of the information system really means, why enterprises can't skip it, and three principles to hold on to when things get fuzzy.

governanceagentsstrategyleadershipgenai
article

Data maltreatment: the quiet tax on the P&L

Data quality is the symptom, data maltreatment is the process that causes it. Its mechanisms, its cost, why incentives keep it alive, and what it does to organisational knowledge and AI.

datagovernanceknowledgestrategygenai
article

Agents that learn on the fly vs. agents you can govern

Self-improving personal agents and industrial agents solve different problems. One learns online, the other learns offline under continuous evaluation. In a regulated company, that difference decides almost everything.

agentsevaluationgovernancellmops
article

Why I'm still a Data & AI leader, and whether that will change

Job titles keep changing, the fundamentals do not. Data feeds AI today, AI produces and shapes data tomorrow. Organisations have to choose between Data & AI and Data vs AI.

leadershipdatagenaioperating-modeltransformation
article

AI is the fifth transformation

In most large organisations, GenAI arrives on top of four other transformations. What I have learnt about delivering value anyway, the battles I avoid, the ones I think are worth fighting, and the people who make it work.

transformationstrategyleadershipgenai
article

When text becomes warehouse data

Language models can finally turn emails, contracts and reports into rows in the data warehouse. Those rows are assertions, not facts, and users need to see that they came from a model.

datagenaidata-warehousegovernance
article