I write this column the way I work with clients. Start from a real question, get the data straight, then say what I actually think.
The research is AI-assisted, and I would rather tell you that than pretend otherwise. I use AI tools to scan the week in association technology, pull the primary sources into one place, and pressure-test every number before it gets near a draft. Once a piece is published, an automated check re-reads it each week and re-opens every citation, so a figure that has gone stale or a link that has rotted gets caught and fixed rather than sitting there wrong.
What the AI does not do is decide what is worth saying. The thesis, the judgment, the willingness to break with the consensus, and the voice are mine. When a piece argues that platform consolidation is a reason to stay composable, that is my read of the field, not a model's. The tools get me to the evidence faster and keep me honest about it. The opinion is the part you cannot automate, and it is the only part worth your time.
If you ever find a claim here that does not hold up, tell me. Getting it right matters more to me than getting it out.
Back to the column, or read the perspectives for the longer arguments behind it.