Issue 29 companion · AI Is Quietly Thinning Your Thinking Muscle
The Think-Then-Tool Protocol
The fastest way to get generic strategy out of a brilliant capability is to open it before you have a view of your own. This protocol keeps the first thought yours: dissect the problem like an executive, form a thesis from your own domain expertise, then bring AI in to attack it. A mind that warmed up gets sharpened. A mind that skipped straight to the answer gets replaced.
This one is for readers of The Signal.
Add your name and The Think-Then-Tool Protocol opens right here, on this page. You also get one sharp idea about AI and leadership every Tuesday, each with a companion like this one.
Founding readers also get The Starting Line, my $47 course, free. One email a week. Unsubscribe any time, and your data is yours - we never sell or share it.
What you are opening
Built for this issue. It follows the argument you have just read, one step at a time.
A live status. It tells you what is still blank, because a half-filled sheet changes nothing.
Saved on your device. Close the tab and come back to it.
Prints clean. Take it into the room on paper if that is how the room works.
The Signal
The Think-Then-Tool Protocol
Issue 29 companion · AI Is Quietly Thinning Your Thinking Muscle
The Think-Then-Tool Protocol
The fastest way to get generic strategy out of a brilliant capability is to open it before you have a view of your own. This protocol keeps the first thought yours: dissect the problem like an executive, form a thesis from your own domain expertise, then bring AI in to attack it. A mind that warmed up gets sharpened. A mind that skipped straight to the answer gets replaced.
The struggle was not a delay in the thinking. It was the gym where the thinking got built.
Nothing answered yet
Part One · Think first — no model open
Everything in this part happens in your own head and your own words. If you cannot fill it in, you do not understand the problem yet, and no capability will fix that for you.
1 · Dissect the problem — the 4-Filter Framework
Break the problem down the way the answer will have to hold up. Four filters, from Thinking at Machine Speed.
The real problem, in one line (the problem, not the symptom)
Decision — what actually gets decided here, and by whom
Audience — who acts on the answer, and what they already believe
Constraints — the money, time, risk, and politics the answer must survive
Output — the form the answer must take (a board paper, a number, a go / no-go)
2 · Write your own thesis first — five bullets, no tool open
Your take, before the machine has one. Infuse your domain expertise, your experience, your hard-won lessons. This is the part AI cannot do for you, and the part that makes its answer worth anything.
3 · Name what you are not sure of
Where is your thesis thin? What evidence is missing? What would change your mind? This becomes AI's brief in Part Two.
The assumptions I am least sure of
The evidence I am missing
Part Two · Then work with AI — the CRIT model
Now, and only now, open the model. Do not ask it for the answer. Brief it the way you would brief your sharpest advisor, in four moves.
The CRIT model — Context, Role, Interview, Task — is from Geoff Woods, author of The AI-Driven Leader.
C
Context
Give it everything: the problem, your five-bullet thesis, your constraints, and what you are unsure of. Paste your dissection from Part One. The more it knows, the less generic it is.
R
Role
Assign the exact expertise you need it to hold. "You are a sceptical FMCG strategy director who has run market entries and watched them fail."
I
Interview
The move most people skip. Tell it to interview you before it answers, so it pulls out what only you know. "Ask me up to eight questions, one at a time, before you give a view."
T
Task
Only now, the ask. Specific, and scoped to the Output you named in Filter 4. "Once you understand, pressure-test my thesis and return a one-page board recommendation."
Steal this prompt
Context: [paste the problem, your five-bullet thesis, your constraints, and what you are unsure of]. Role: You are a [specific expert] who has done this many times and is paid to be right, not agreeable. Interview: Before you answer, ask me up to eight questions, one at a time, to get what you are missing. Task: Once you understand, [the specific task] and return it as [the output you named in Filter 4].
Part Three · Let it attack — the adversarial pass
You have a thesis and a briefed advisor. Now turn it hostile. This is where the sharpening happens: the model hunts the holes, and a warmed-up mind closes them. This is the Adversarial mode from the course, aimed at your own thinking.
Steal this prompt
Attack the thesis above. Find every assumption that does not hold, every piece of data I am missing, and build the strongest counter-case to my position. Argue why I am wrong. Do not reassure me, and do not soften it.
What it found that I had missed — and what I now change
The order that keeps the thinking yours
I dissected the problem before I opened anything
I wrote my own five-bullet thesis first
I briefed with Context, Role, Interview, Task
I let it attack my thinking, not do it for me
The rule
If you cannot write the five bullets, you do not understand the problem yet. Think first. Bring AI in to attack, not to author.
The Signal
One sharp idea about AI and leadership, every Tuesday - with a companion like this each issue. Not subscribed yet? Get The Signal here.
VZN Studios · Your data is yours - we never sell or share it.
Join Our Free Trial
Get started today before this once in a lifetime opportunity expires.
