Source: https://x.com/karpathy/status/2037921699824607591 | https://x.com/karpathy/status/2049907410303865030 "Drafted a blog post → LLM argue the opposite" — 31k likes
Core Principle
Don't ask what the LLM thinks. Ask it to simulate what a diverse group of experts would argue.
LLMs are trained to please. A direct question gets a sycophantic answer. A simulation request gets a distribution of real perspectives — including the uncomfortable ones.
Karpathy's method: write a draft → ask LLM to argue the strongest possible opposite position → synthesize a better view.
The 4 Simulator Modes
Mode 1: Expert Debate Panel
Best for: technical decisions, architecture choices, research directions
Simulate a structured debate between these 3 expert personas on [TOPIC/DECISION]:
Expert A: [most optimistic / pro position]
Expert B: [most skeptical / con position]
Expert C: [pragmatic outsider / unexpected angle]
For each expert:
- State their core argument in 3 sentences
- Cite 2 specific examples or data points they'd use
- Identify what they'd say is the FATAL FLAW in the opposing view
After the debate, synthesize: what's the strongest hybrid position that survives all three critiques?
Topic: [YOUR_TOPIC]
My current position: [YOUR_DRAFT_VIEW]
Mode 2: Steel Man the Opposite
Best for: before publishing, before committing to a decision
I'm about to [ACTION / PUBLISH / DECIDE]:
[YOUR PLAN OR DRAFT]
Steel man the strongest possible argument AGAINST this. Be merciless.
Don't hedge. Don't say "while this has merit...".
Argue as if you genuinely believe the opposite and need to convince a skeptical expert.
Then: what would it take to make my original position survive this attack?
Mode 3: Pre-Mortem Simulation
Best for: project planning, product launches, major decisions
Imagine it's [DATE 6 MONTHS FROM NOW] and [YOUR PROJECT/PLAN] has failed completely.
Simulate 3 different failure modes — each from a different root cause:
1. Technical failure: what went wrong in the implementation?
2. Strategic failure: what assumption proved wrong?
3. Execution failure: what human/process error occurred?
For each: describe the specific sequence of events that led to failure.
Then: what early warning signals would have been visible by [DATE 1 MONTH FROM NOW]?
Mode 4: Socratic Drill
Best for: testing your own understanding of a concept
I believe I understand [CONCEPT].
Ask me 5 increasingly hard Socratic questions to test my understanding.
After each answer I give, probe a potential gap or inconsistency.
At the end, grade my understanding on a scale of 1-10 and identify my biggest blind spot.
Start with question 1.
Anti-Sycophancy Patterns
Always add one of these to prevent the LLM from capitulating to your framing:
"Do not soften your critique because you think I want to hear it.""If my position is clearly wrong, say so directly.""Ignore that I wrote the original draft — evaluate it as an anonymous submission.""Your job is to find the fatal flaw, not to validate."
Decision Framework
Before any major decision:
1. Write down your current position (1-2 paragraphs)
2. Run Mode 2 (Steel Man the Opposite)
3. Update your position based on the strongest critique
4. Run Mode 1 (Expert Debate) to get broader perspective
5. Make the decision — now you've genuinely stress-tested it
Total time: 15-20 minutes. Quality of decision: dramatically better.
When NOT to Use This
- For factual lookups (just ask directly)
- For code generation (use agentic engineering instead)
- When you've already decided and need to execute (move forward, don't ruminate)
Workflow
Belongs to workflow: anti-bias decision-making (any important judgment)
| Position | Upstream | Downstream |
|---|---|---|
| Entry point | Triggered directly when the user faces an important decision | karpathy-understanding-first(verify whether you truly understand) |
Full chain:llm-simulator → understanding-first → system-prompt-learning
Also used in the monthly checkup workflow to stress-test monthly conclusions.
Prompt Contract
Simulate a structured debate between 3-5 expert personas with conflicting incentives on <DECISION/TOPIC>. Each persona must: state their core argument, cite specific evidence, identify the fatal flaw in opposing views. After the debate, synthesize the strongest hybrid position that survives all critiques. End with: key assumptions I must verify myself.
Verification Checklist
- At least 3 roles, with interests/positions that conflict with each other
- Each role provides specific arguments (not vague generalities)
- Each role points out the other side's fatal flaw
- The综合 conclusion withstands attacks from all roles
- Lists the assumptions "I must personally verify"
- No tendency to flatter the user's original position