One signal-dense note when a new episode lands. No noise.
Control Is the Inverse of Trust - Part 4: Natural Value Flow over Artificial Control
August 5, 2026
Three operators — Chris Carolan, Trisha Merriam and Erin Wiggers — take Natural Value Flow over Artificial Control out of the abstract and run it against systems they actually operate: a live HubSpot go-live, an agent memory system, and a $100K AI transformation contest whose rules quietly reward the wrong thing. The through-line the conversation actually found is that control and trust are one dial, not two — you need less control of a system exactly as fast as you build reasons to trust it — and that the highest-leverage move is not removing gates but moving where you point them: at the goal and the North Star, not at the build. Chris tells the origin story of Value-First for the first time on the show, and the episode closes on the most ordinary form of artificial control there is: renaming an industry term inside your own CRM.
Moments from this episode
Key takeaways
Control and trust are one dial, not two. The more you trust a system, the less control you need over it — so the work is building reasons to trust, not writing rules.
Give up artificial control, not all control. The operative word is 'artificial': the control worth removing is the control that was never doing any work.
Define the hard lines and people will hand over far more than you expect. Naming what the AI will never do is the fastest way to make a human comfortable letting it act.
Scope the goal, not the build. Telling AI which filters to use takes every better option off the table before it has a chance to help you.
If you can't say what a bad answer would look like, you're not prompting — you're asking for a compliment. Frame a prompt as a hypothesis and the guardrails install themselves.
State your own thinking first, then ask what you're missing. Claiming a position frees the model to leave it, instead of validating the first idea you had.
It cannot be AI-first or human-first — it's wherever the natural value is. That is what makes 'over' a direction rather than a rejection.
The measurement is what does the controlling. When calls are the number, an hour that solves the problem is punished and five minutes that solves it is impossible.
You can run a real AI transformation with zero AI builds. Removing the spreadsheets and unifying context is the transformation; the agents are what becomes possible afterward.
Not enough time to learn the business isn't scope creep, it's inadequacy — and the shortcuts that fill the gap are always the same industrial defaults.
Point control at the North Star — the values and beliefs — and you no longer need the scope of the build figured out up front.
The most ordinary artificial control is renaming something. Invent your own term for an industry standard and you spend money on inbound data you then have to fight.
Keep learning
You just watched the work. Now build it.
The next step is doing it yourself — live, with people on the same path.
One signal-dense note when a new episode lands. No noise.