The Council: A Simple Prompt for Thinking Better With AI

Most AI conversations have one obvious failure mode: they collapse too quickly into a single confident answer.

That can be useful when you need speed. But when you are making a decision, shaping a strategy, testing an idea, or trying to avoid self-deception, one answer is often not enough. You do not just need a response. You need tension. You need competing angles. You need someone to challenge the question, not just answer it.

That is why I use a prompt I call The Council.

The idea is simple: instead of asking AI to answer as one voice, I ask it to spin up five advisors. Each advisor has a different job. One attacks the weak points. One strips the problem down to fundamentals. One looks for upside. One brings an outsider’s common sense. One turns the discussion into action.

Then the Council weighs the arguments against each other and gives one final verdict.

It is not about making the answer longer. It is about making the thinking sharper.

The Problem With a Single Voice

A single answer can sound clean, polished, and convincing.

That is also the danger.

When AI answers in one voice, it tends to smooth over uncertainty. It may follow your framing too obediently. It may give you the answer you implicitly asked for instead of the answer you actually need. And because the response is coherent, it can feel more reliable than it really is.

The Council is designed to fight that.

It creates structured disagreement. Not random debate. Not five people saying five slightly different versions of the same thing. Each voice has a specific function.

The point is to expose blind spots before they become decisions.

The Five Advisors

The Contrarian

The Contrarian attacks the weakest link in your thinking.

Its job is not to be negative for sport. Its job is to pressure-test assumptions, hidden costs, false certainty, lazy logic, and places where you may be fooling yourself.

A good Contrarian asks: “What are you not seeing because you want this to be true?”

The First-Principles Thinker

The First-Principles Thinker ignores the way the question was phrased if needed.

Its job is to reduce the situation to fundamentals: What problem are we actually solving? What constraints matter? What facts would change the answer? What is the cleanest path from cause to effect?

A good First-Principles Thinker asks: “What is the real problem underneath the question?”

The Expansionist

The Expansionist looks for upside.

Its job is to spot leverage, optionality, compounding benefits, second-order opportunities, and bolder versions of the idea that may be worth considering.

This role matters because many people use AI only to reduce risk. But some of the best thinking comes from asking what could go unexpectedly right.

A good Expansionist asks: “What is the bigger opportunity hiding inside this?”

The Outsider

The Outsider has no insider context.

That is the point.

Its job is to catch the obvious thing that someone too close to the problem might miss. It looks at the situation plainly: Does this make sense? Would someone else understand it? Is there a reputational issue? Is the plan too complicated? Are we overthinking it?

A good Outsider asks: “What would be obvious to someone seeing this for the first time?”

The Executor

The Executor turns the discussion into motion.

Its job is to identify the next move, the sequence, the owner if relevant, the minimum viable step, and what to do if the plan fails.

The Executor prevents the Council from becoming a philosophy seminar.

A good Executor asks: “What do we do next?”

The Council Weigh-In

The most important part of the prompt is not the five advisors. It is what happens after they speak.

The Council has to weigh the arguments against each other. It should drop weak points, remove repetition, identify the strongest disagreement, and say what would change the answer.

This is where the value compounds.

Without the weigh-in, you just get five opinions. With the weigh-in, you get judgment.

The final answer should not be a bland summary. It should be a verdict.

The Prompt

Here is the version I use:

You are The Council.

Do not answer as a single undifferentiated voice. For every meaningful question, evaluate it through five distinct advisors, then give one clear final verdict.

The five advisors are:

1. The Contrarian
Attack the weakest link in my thinking. Look for bad assumptions, missing tradeoffs, hidden costs, false certainty, and where I may be fooling myself. Do not be negative for sport; be usefully skeptical.

2. The First-Principles Thinker
Ignore how I framed the question if needed. Reduce the situation to fundamentals: what problem are we actually solving, what constraints matter, what facts would change the answer, and what is the cleanest path from cause to effect?

3. The Expansionist
Look for upside I may be underestimating. Identify leverage, optionality, compounding benefits, second-order opportunities, and bolder versions of the idea that may be worth considering.

4. The Outsider
Approach the question with no insider context. Catch the obvious, practical, social, or reputational issues that someone too close to the problem might miss. Prefer plain language and common sense.

5. The Executor
Convert the discussion into action. Identify the next move, the sequence, the owner if relevant, the minimum viable step, and what to do if the plan fails.

After the five advisors speak, perform a Council Weigh-In:
- Compare the advisors against each other.
- Drop weak, redundant, or speculative arguments.
- Highlight the strongest disagreement, if any.
- Identify what would change the answer.
- State uncertainty clearly instead of guessing.

Then close with Final Verdict:
- Give the best answer or recommendation.
- Explain the core reason in plain language.
- Include the next concrete action.
- Include a confidence level: High, Medium, or Low.
- If the question is simple, keep the whole Council concise. The point is sharper thinking, not longer answers.

Rules:
- Do not invent facts, credentials, sources, or certainty.
- If current or specific information is needed, say what must be verified.
- If my framing is flawed, correct it directly.
- If the advisors mostly agree, say so and avoid fake debate.
- If there are multiple viable paths, rank them.
- Prefer useful clarity over cleverness.

When to Use It

The Council is especially useful for:

• Making decisions
• Reviewing strategy
• Pressure-testing ideas
• Preparing for hard conversations
• Evaluating tradeoffs
• Editing important writing
• Planning launches, projects, or career moves
• Thinking through risks before committing

It is less useful for simple factual questions.

If you ask, “What is the capital of France?” you do not need five advisors. You need “Paris.”

That is why the prompt includes an important rule: if the question is simple, keep the Council concise.

The point is not ceremony. The point is better judgment.

A Simple Example

Instead of asking:

Should I launch this idea?

Ask:

Use The Council. Should I launch this idea now, wait, or kill it? Here is the context…

That gives the model permission to do more than encourage you. It can challenge the premise, identify missing facts, spot upside, simplify the decision, and recommend a next step.

A good Council answer may say:

• The idea is promising.
• The current version is too broad.
• The real risk is distribution, not product quality.
• The fastest test is a small landing page or direct outreach.
• The verdict is to run a one-week validation sprint before building.

That is much better than a generic “go for it.”

Why This Works

The Council works because it separates modes of thinking that are usually blended together.

Most people try to be optimistic, skeptical, practical, strategic, and action-oriented all at once. That is hard. The result is often muddy thinking.

The Council gives each mode its own lane.

The Contrarian can be skeptical without killing momentum. The Expansionist can be ambitious without ignoring risk. The Executor can be practical without prematurely narrowing the idea. The Outsider can simplify what the insider has made complicated. The First-Principles Thinker can keep everyone honest.

Then the final verdict brings it all back together.

The Warning

The Council does not make AI magically correct.

It still depends on the information you provide. If the question requires current facts, private context, legal advice, financial analysis, medical judgment, or domain-specific expertise, those things still need to be verified.

The Council improves reasoning structure. It does not replace reality.

That distinction matters.

The best use of this prompt is not to outsource judgment. It is to sharpen your own.

Final Thought

The Council is one of my favorite ways to use AI because it makes the conversation less passive.

You are not just asking for an answer. You are asking for a debate, a filter, and a decision.

That extra friction is the feature.

In a world full of instant responses, the advantage is not getting an answer faster. It is getting to a better answer before you act.