Big Tech Accountability
Cluster · workshop-as-shared-build

What You Take Home From a Workshop

By Michael Polzin · July 1, 2026 · Human-in-the-loop, tightly qualified.

A workshop is not a lecture you attended. It is a build you left with. This page is the inventory: what actually goes home with you, and what does not.

The three things in your bag

When the day ends, you leave with three concrete artifacts. Not a certificate. Not a pep talk. Three things you can open on a Tuesday and use.

1. Your build. A running, inspectable configuration that you assembled in the room, with your own inputs, on a schema we walk through step by step (Class C, configuration and integration). It is yours: your files, your names, your constraints. If you cannot open the folder afterward and read every gate, we did not finish the workshop.

2. Your notes. A written log of the decisions you made, in the order you made them, with the reason recorded next to each one. This is the ledger habit in miniature (Class B, code and inspection): the point is not the notes themselves, it is that next month, when you disagree with your own past decision, you can find the evidence class you attached to it and either upgrade the claim or drop it.

3. Your loop practice. One rehearsed pass of the observe, model, act, review cycle we call the LOOP. You will have run it at least once on a real problem you brought in, with a human in the loop the whole way (Class A, empirical in session). One pass is not mastery. It is a proof, in your own hands, that the loop is a thing you can do, not a slogan you can quote.

What we are careful not to promise

You will not leave with a certification, a credential, or a claim that you now practice active inference. Active inference is a technical framework in cognitive science (Class E, expert citation: Parr, Pezzulo, Friston, 2022); a one-day session is an introduction to a working posture, not initiation into the mathematics. If a workshop tells you otherwise, ask for the receipt.

You will not leave with a fix for anything clinical. Our work is trauma-informed as a facilitation posture, meaning we design the room so people can think under pressure. It is not therapy, it is not clinical care, and it does not treat any condition. If you need clinical support, we will help you name that and route you to it.

You will not leave with a guarantee that your build will hold in production. It is an early draft, made in a day, by you, with our help. It is meant to survive contact with reality, and we expect it to change. What we do promise is that you leave knowing where the fences are: which parts you tested in the room, which parts are still marked Class U, unverified, and which parts you should not ship until you have your own evidence.

How this fits the larger frame

The workshop is a shared build, not a shared belief. UNI, the program the workshop teaches from, is a working hypothesis on an attainable path toward General Natural Intelligence: a natural, active-inference approach whose evidence is growing, evidence-classed, and tested in the open. Do not take the claim on faith. Test the build, inspect the gates, and help us find where it fails (Class F, falsifier present).

Adjacent to our program, Themesis publishes long-form work on active inference and cognitive architecture. We link them as a factual pointer to the wider field, not as an endorsement of our workshop: their scholarship is theirs, our build is ours, and readers should read both on their own terms.

A short note on why we call it LOOP

One word about the naming. We do not use the word that rhymes with LOOP and starts with L E. We chose LOOP because the point is the return, not the jump: observe, model, act, review, then loop back with what you learned. The name has to do the work of reminding you to come back around. If you are curious about the naming choice itself, we wrote it up separately.

What to do next