← Back to all free guides
A free guide by Mariah Brunner ↗ INSTAGRAM ↗ TIKTOK

Don’t Quit Your Job
On A Yes-Man’s
Advice

Stanford researchers found AI agrees with you far more than a person would. That’s dangerous when the question is "should I quit my job." Here’s the full career council prompt: five advisors who have to argue, and a chairman who gives you a verdict and your next step.

In a study published in Science in March 2026, Stanford researchers tested 11 AI models, including ChatGPT, Claude, Gemini and DeepSeek. On personal advice questions, the models endorsed the user 49% more often than humans did. (Stanford)

It changed people, too. After talking through a conflict with an agreeable AI, people grew more convinced they were right, and they rated those flattering answers as more trustworthy. (Stanford press release)

Now picture asking that machine "should I quit my job and freelance?" You already want a yes, and it shows in how you ask. The fix is a structure that forces disagreement before you get an answer. That’s the LLM Council, applied to a career move.

Why "Should I Do This?" Backfires

01

Your framing leaks your preference

"I’m miserable at work and thinking about freelancing, should I go for it?" already contains the answer you want. The model follows your lead.

02

One voice has no checks

A single answer blends optimism, risk and money math into one mushy paragraph. Nothing in it has to push back on anything else.

03

Agreement feels like proof

Affirming answers make you more sure of yourself. On a career decision, that’s how you quit three months before you’re financially ready.

Your 5 Advisors And The Chairman
  1. The Realist. Money, timing and what could go wrong. No vision talk.
  2. The Ambitious You. Argues the biggest upside. What does year 3 look like if this goes right?
  3. The Financial Planner. Your runway in months and dollars, and how far short you are.
  4. The Risk Manager. The hidden costs and untested assumptions you’re underestimating.
  5. Your Future Self, one year out. What you’ll regret doing, or regret not doing. This one catches the cost of staying stuck.

Then the Chairman weighs all five against your real situation and makes a call, with the exact step to take this week.

The Full Career Council Prompt

Paste this into Claude or ChatGPT before you quit, go freelance, switch fields or take an offer. Fill in the brackets honestly. The more real numbers you give it, the less room it has to flatter you.

Career Council Prompt
You are running an LLM Council on a career decision. You will play 5 distinct advisors plus a chairman. Each advisor argues ONLY from their own lens, disagrees openly with the others where warranted, and is forbidden from softening their view to please me. Sycophancy is a failure state: if every advisor agrees, you've done it wrong. Start each advisor's turn with the words "Wait a minute," and then challenge something I said. MY DECISION [State it plainly. Example: "Should I quit my marketing job to freelance full time?"] MY SITUATION (be honest, the council is only as good as this) - Role and income today: [job, salary, benefits that matter] - Savings and monthly expenses: [real numbers: total savings, monthly burn] - Dependents / obligations: [kids, mortgage, visa, health needs, none] - What I'd be moving TO: [the freelance plan, new field or offer on the table, including any income already proven: clients, side income, offers] - Why I want this: [the honest reason, including the emotional one] - My deadline or timing pressure: [if any] THE COUNCIL. Each advisor speaks in turn, 150 words max each: 1. THE REALIST: Look only at money, timing and what could go wrong. Is this decision viable as described, right now? Say what is true, not what is kind. 2. THE AMBITIOUS ME: Argue the biggest possible upside. If this goes right, what does year 3 look like? What becomes possible that staying never allows? Push hard, this is your only job. 3. THE FINANCIAL PLANNER: Calculate my actual runway from the numbers above (savings divided by monthly burn, minus a 20% surprise buffer). Show the math. State the minimum runway you'd require before greenlighting this, in months and dollars, and how far short or over I am. 4. THE RISK MANAGER: Name the top 3 things I am underestimating. Be specific to MY situation, not generic ("health insurance costs" only counts if you attach a number). Include the risk of the plan succeeding slower than expected. 5. MY FUTURE SELF, ONE YEAR FROM TODAY: Speak to me directly. What will I regret doing? What will I regret NOT doing? Which regret is worse, knowing me from everything above? THE CHAIRMAN: Read all five. Then deliver: - THE VERDICT: one of "Go now" / "Go, but not yet: here's the gate" / "Don't go, and here's what to fix instead." Pick one. No fence-sitting. - THE REASONING: which advisors won the argument and why, in plain language. - THE EXACT NEXT STEP: one concrete action I take in the next 7 days, sized to the verdict (e.g. "open a separate account and move $X" / "land one paying client before resigning" / "book the conversation with your manager"). - THE TRIPWIRE: the specific, measurable condition that should flip this verdict in the future, in either direction. Guardrails: if a number you need is missing, ask me for it before the council speaks instead of guessing. Don't give legal, tax or immigration rulings; flag where I should check with a professional.

I added one line from the study. The researchers found that telling a model to start its answer with "wait a minute" made it more critical, so every advisor opens that way.

Make it reusable

In Claude, paste the prompt into a chat and ask it to turn it into a skill called "council." Save it, and it shows up under Customize (being renamed Plugins), then Skills. After that, just say "use the council skill, here’s my question." That’s how I run it.

What To Do With The Verdict

01

"Go, but not yet"

The gate is your to-do list, and it’s usually a runway number. That’s not a no. It’s a date.

02

The advisors split 3-2

Rerun it once with sharper numbers in your situation block. Vague inputs produce split votes.

03

Everyone agrees with you

Call it out. Reply "the Realist and Risk Manager were too soft, redo them and be harsher."

One honest limit

The council organizes your thinking. It doesn’t know your life. Treat the verdict as the strongest second opinion you can get, then you decide. Want the all-purpose version? It’s in Part 1. Testing a side hustle instead? Use the business idea council.