Placement in our AI programs is by demonstrated readiness
Why we place people by what they can show rather than by what they choose to pay for.
Every certification has a moment where it decides what it is actually selling.
It happens at the point of enrolment, and it is usually invisible. Someone arrives, looks at three tiers, and picks one. If the only thing standing between them and the top tier is a payment, the certification has just told you what it values. It values the payment.
That is the ordinary way it is done, and it is not done cynically. Self-selection is easy to build, easy to explain, and it converts well. It also quietly transfers the hardest question in the whole process onto the person least equipped to answer it, which is the applicant deciding how good they are at something they have not been assessed on yet.
We place people instead. You apply once, and we decide where you sit.
What we are actually looking at
The application asks what you have built, taught, advised on, or been accountable for. Not what you have read, not what you have completed, not how long you have been interested in AI. What you have done, and what happened as a result.
Three levels, and the difference between them is not difficulty. It is the scope of what you are responsible for.
Catalyst is for the person building their own judgment and wanting a framework underneath it. You are using AI in real work, you have opinions formed from experience, and you want structure that holds those opinions up rather than replacing them.
Architect is for the person already teaching AI and improvising the structure. You are in front of rooms. You are answering questions on the spot. What you are missing is not confidence, it is a standard you did not have to invent yourself on the drive there.
Steward is for the person who will hold the standard for other people. You are not just delivering, you are deciding what good looks like and being accountable when it is not.
Read those again and notice that none of them is a beginner tier or an advanced tier. They describe different jobs.
The failure this is designed to avoid
Two failures, actually, and they pull in opposite directions.
The first is the one everyone recognises. Someone pays for the senior tier, sits in a room they are not ready for, and leaves with a credential that describes a capability they do not have. That is bad for them, because the first time it gets tested in public it fails. It is worse for everyone else holding the same credential, because the credential now means less than it did before they walked in.
The second failure is quieter and we take it just as seriously. Someone with fifteen years of real accountability puts themselves in the entry tier because they are not sure they count, sits through foundations they built their career on, and concludes the whole thing was not for people like them. Impostor syndrome is not a personality quirk. In a self-selecting system it is a pricing mechanism, and it consistently underplaces the people who would be best at this.
Self-selection gets both of these wrong in the same way. It asks people to assess themselves and then charges them for the answer.
What placement costs us, and why we pay it
Placement is slower. Every application is read by a person. We go back with questions when something is unclear. Some people are placed higher than they applied for, which means a conversation about a cohort that is not the one they were looking at. Some are placed lower, which is a harder conversation and we have it anyway.
None of that scales the way a dropdown menu scales. We accept that, because the alternative is a credential where the level tells you about someone's budget rather than their practice.
There is a version of this that goes too far, and it is worth naming. A process that assesses so thoroughly that nobody gets in is not rigour, it is gatekeeping wearing rigour's clothes. We are not building an exam. We are building a placement that takes about as long as a careful read and a short conversation, and then gets on with it.
How it works in practice
You apply once. There is no separate application per level.
We read what you sent and place you. If that placement is Catalyst, the cohort starts 15 September and applications close 11 September.
If your experience places you at Architect or Steward, you go straight onto the list for that cohort and we come back to you directly. You do not reapply. You do not pay to hold a place. The list exists so that people who are already operating at that level are not asked to sit through the level below to prove it.
Nobody buys their way up. That is the entire point, and it is the reason the credential means something when someone puts it next to their name.
The wider argument
Capability is not authority.
Being good with AI does not make you qualified to teach it. Those are different skills, and the gap between them is where most of the harm in AI education currently sits. Someone confident and capable in their own use, standing in front of thirty people, teaching workflows without the framework to know which of them will fail badly and for whom.
A credential that lets people self-select their level does nothing about that gap. It widens it, because it hands the strongest signal to whoever felt boldest on the day.
We would rather do the slower version.
Applications for Catalyst close 11 September. The cohort starts 15 September.
