Blog
Thoughts and insights from practice.
On AI strategy, adoption, and getting artificial intelligence out of slide decks and into production. Written from practice, not from keynotes.
The biggest mistake banks make with AI
Banks go looking for places to put AI instead of solving a real need. Why the low-hanging-fruit pilot dies first, and what to back instead.
Sovereign AI and Europe's dependency
Availability, pricing, and now geopolitics. Why companies and Europe should think about sovereign AI infrastructure before they need it.
The closed-loop organization
Prompting agents is already obsolete. What the closed-loop organization is, why middle management disappears, and what it does to cost structures.
The epiphany problem in AI adoption
You can't transfer the understanding of agentic AI with slides. Everyone needs their own epiphany. Why productivity is a cliff, not a bell curve.
Why AI adoption fails in companies
Buying licenses is easy. Changing the way a company works is the hard part. What separates companies that actually make money with AI from those with an expensive subscription.
Takeaways from The Banking Scene Brussels
Four honest takeaways from Brussels: the agentic gap, the System 1 problem, what Revolut gets right, and why AI amplifies expertise rather than democratizing it.
One paying customer
Don't bring me a use case. Bring me one paying customer. The brutal filter that kills 80% of AI ideas and lets the one that actually ships through.
Agentic AI in banking: between hype and bank-grade reality
AI agents will change banking. But not the way conference keynotes promise. What it means to deploy agents in an environment where mistakes are measured in money and trust.
What intelligence analysis taught me about working with AI
Five years in the defence community taught me to work with uncertainty, evaluate sources, and expect disinformation. Exactly the skills everyone working with AI needs today.