AI costs: why this isn't a bubble but a bill
Stock correction, export controls and rising token costs: AI isn't a bubble but a bill. What does that mean for your governance?
Blog
Insights and perspective
I regularly share insights on AI, strategy and the choices organisations need to make now to stay relevant.
AI hallucinates 86% of the time and solves maths problems that stumped humans for decades: how can both be true?
AI hallucinates 86% of the time yet cracks Erdős problems. Not contradictory — this is the absent-minded professor in your laptop, and what to do about it.
Withdrawal or cancellation: why the new button is more than a compliance fix
The withdrawal button law covers fourteen days — a compliance mistake extends that to 12 months. Why the best exit flow beats the legal minimum.
There are four kinds of AI users in your organization. You probably treat three of them wrong
Every organization has four kinds of AI users: the pioneer, the craftsman, the builder and the searcher. They respond in opposite ways to the same levers, so treat everyone the same and you treat three out of four wrong. The full model is in a free whitepaper.
AI vendor lock-in: why the kill switch is your problem
AI vendor lock-in is not about which model you choose — it is about where your agents run. On June 12, 2026, the White House showed that access to an AI model can disappear with a single government letter. Sovereignty is not a model choice. It is an architecture choice.
AI supervision for insurers: the division of tasks is not your problem
The insurance sector wants regulatory clarity on AI oversight. But the real brake on AI adoption isn't a supervision problem — it's a knowledge problem.
Open source AI: from smart alternative to governance necessity
Costs are rising, a government shut down an AI model, and the hardware became affordable. Why open source AI is no longer optional for leaders — it's insurance.
AI Is Not Always the Answer
95% of AI pilots deliver zero P&L impact. Technology doesn't fail — the approach does. Why asking 'what can AI do here?' is the wrong first question.
Voice AI customer service sounds more human than your staff
A restaurant helped me brilliantly. Friendly, smooth, no wait. Only after hanging up did I realise I had spoken to an AI. That unsettled me. Not because it went badly, but precisely because it went so well.