LLMs Predict Language. Humans Construct Meaning.

Large language models produce fluent language in seconds. Piaget helps explain why understanding still has to be constructed.

Quantum Computing

You can already use a quantum computer

Quantum computing still feels distant, yet anyone can already write, simulate and run simple quantum programs. The software ecosystem is arriving before the hardware matures.

AI in Practice

Looking Past the Model

ChatGPT Work and Kimi K3 made me look beyond the engine. Increasingly, the meaningful product may be the environment in which a model can actually work.

AI Strategy

Is there really one AI economy?

Over the past few weeks I've found myself reading widely differing takes on AI. Some see the biggest speculative bubble in history. Others see the foundations of a new industrial era. Perhaps they're simply looking at different parts of the same landscape.

AI Strategy

Why I Keep Choosing Proprietary Products

Why do I keep choosing proprietary products when open-weight models promise more control, flexibility and independence? The answer may lie less in principle than in what is actually usable, affordable and reliable today.

Quantum Computing

Quantum computing is more than qubits

A Dutch quantum computer just reached its first customer, not for its power to solve problems yet, but to learn what building on this new kind of computer means.

Sovereignty

Compute Is the Entry Ticket, Not the Game

Compute is necessary but not sufficient. A LinkedIn debate sharpened an argument about European AI infrastructure into a longer chain: energy to demand.

Sovereignty

A Serious European Bet on AI Compute

VOLT's proposed Rotterdam AI campus would be among Europe's largest, if built. The real scale, and why Dutch soil isn't the same as durable Dutch control.

AI Governance

A License to Prompt

A vocal minority told me I wasn't entitled to use a frontier model for a "simple" diagram. It reveals a split running through centuries of regulation: license the person, or license the maker.

AI in Practice

Fable Is Back. Brilliant, and Expensive.

Fable is the best model I have, and the heaviest. It drains my usage faster and costs more past the limit. On the price of reaching for the strong tool.

AI Governance

Vanta proves the lock is installed

Vanta makes vendor trust continuous instead of episodic. What it automates is real. What it still can't do is tell you whether that's enough.

AI Governance

How safe is that, actually

A vendor I vouched for got challenged twice in one week. I couldn't answer with more than trust, so I built a method to climb past it.

AI Strategy

What Clients Mean When They Say Sovereignty

Sovereignty has become one of the most common requirements in AI and digital procurement. The problem is that the word usually describes several different concerns at once.

Identity

SimpleX: The Messenger Without Accounts

No phone number. No username. No directory. Just a secure connection between two people.

AI in Practice

Is Your Site Agent-Ready?

Cloudflare quietly shipped a scanner that scores how well your website serves AI agents, and hands you a fix list your coding agent can act on directly.

AI in Practice

The €60,000 LLM Well

Running AI on your own hardware sounds like freedom. The LM Studio demo at WWDC26 makes you calculate what that actually costs, and what it buys you.

Identity

Bearer: When Possession Is the Point

A train ticket added to an app that didn't ask my name cracked open a distinction I hadn't seen clearly before: two models of access, and they explain more than I expected.

AI Governance

Fable 5 Is Gone. The Switch Was Always There.

Yesterday I was using Fable 5. Today it's gone. The US government's export control directive on Anthropic's models is the first time I've felt AI dependency as something physical.

AI in Practice

The field is the lab

Two projects, one platform, two completely different ideas about what research means. I keep noticing this pattern. It says something about what AI tools actually are.

Identity

The App Store Is Not the Argument

Criticising the EU's digital identity wallet for requiring an Apple or Google account mistakes the delivery mechanism for the dependency. The real question is elsewhere.

AI in Practice

The same problem, thirty years later

From early websites to AI assistants, the core challenge has never changed: how do you make what an organisation knows available to the people who need it?

AI Governance

The Poisoned Library

When an AI agent trusts a tool server the way an app trusts a package, the supply chain attack surface moves with it. This is not a new problem wearing new clothes.

AI Strategy

The Layer AI Is Moving Into Now

AI started in the operational layer. It is now working its way into coordination, prioritisation, and strategy. That shift raises a question most organisations have not yet asked.

typography

A font is not a file

For thirty years I thought of fonts as objects. A missing font on a working day, and a workaround from my wife, changed that mental model completely.

Tools

Back in the Search Arena, This Time With Vectors

Three years after writing about site search, I rebuilt mine from scratch: Algolia for full-text speed, a vector database for semantic retrieval, and a RAG pipeline that made both layers useful.