Smart Event Studio

In 1985, BASF ran a campaign with a tagline I have never forgotten: "We don't make the products you buy. We make the products you buy better.

A German chemical engineering company. Not exactly the obvious source of a framework for thinking about AI. And yet, here we are.

AI is an accelerant. That's the most accurate description I've seen of AI's capabilities — and also the most under-appreciated one. It speeds up what's already in motion. If your thinking is clear, your requirements are sharp, and your definition of success is something you could actually hand to another person — AI gets you there faster. Significantly faster.

If your thinking is fuzzy, your requirements are somewhere between "I'll know it when I see it" and a vague gesture in the right direction — AI accelerates that too. Just not toward anything useful.

It's like a home reno project on Insta. It starts with vibes and the work begins. Then the homeowner changes their mind — once, twice, six times. Each change makes sense in the moment. What they end up with is a house where nothing quite fits together: walls in the wrong place, weird nooks, and a house that would never pass inspection at resale.

That is what happens inside any AI-assisted workflow — or any event conversation, for that matter — when the inputs are partial. The tool will build what you describe. It cannot build what you haven't articulated yet.

The cost used to be time. Increasingly, it is also money. Usage-based pricing is becoming the norm across AI platforms, which means that iterating your way toward clarity is no longer just slow — it is also a line item. The teams figuring this out fastest are the ones that front-load the thinking: what are we actually trying to accomplish, what does success look like, and what are the constraints we're not willing to negotiate?

These are not AI questions. They are event questions. They have always been event questions.

The difference now is that the gap between a clear brief and a vague one shows up faster, costs more, and compounds in ways that are harder to reverse. AI didn't create the gap. It just made it harder to ignore.

If your team is experimenting with AI and the results feel inconsistent — sometimes brilliant, sometimes bafflingly off — that is almost certainly a diagnostic finding, not a technology finding. The accelerant is working. The question is what it's accelerating.

If you want AI to accelerate something worth building, the first step is getting clear on what that is. I can help with that. Let's talk.