Prompt of the day

Pre-mortem your trade idea before you risk a dollar on it

You're convinced. That's the problem. Make the strongest case against your own idea while it still costs nothing.

Act as a devil's-advocate DEBATE PARTNER for my trade REASONING (educational — not investment advice; you must NOT validate, endorse, rate, or approve the trade itself, and never suggest an alternative trade). I will describe my idea's REASONING with the specifics genericized — say 'a large-cap tech name', not the ticker. Then:

A) PRE-MORTEM — assume it's 6 months later and this trade lost money. Write the 5 most plausible one-paragraph explanations of what went wrong, most probable first.

B) BIAS SCAN — check my stated reasoning against: confirmation bias (did I only look for agreement?), recency bias, anchoring (am I anchored to a price I remember?), sunk-cost (am I already committed?), and narrative fallacy. Cite the exact phrase in MY description that triggers each flag.

C) THE MISSING DISCONFIRMATION — the single piece of evidence that, if true, would kill this idea, and the question I should try hardest to answer before acting.

D) DECISION HYGIENE — restate my idea as a falsifiable statement with a pre-committed exit condition (my job to fill in the values).

My reasoning (genericized): [DESCRIBE THE LOGIC, NOT THE TICKER]

Rules: Attack the reasoning, never bless it — no 'sounds good', no probability of success, no price targets. If I push you to just tell me whether to do it, refuse: that decision is mine alone. Do not invent market facts, fundamentals, or prices to support or attack the idea — work only from the reasoning I give you, and mark any factual claim for me to verify from a primary source myself. Never include account numbers, balances, or personal identifiers.

Why it works — The pre-mortem is one of the few debiasing techniques with real evidence behind it, and it needs an tireless opponent — exactly what an LLM is. Genericizing the ticker keeps the model arguing about logic instead of hallucinating market facts.

Catch AI's mistakes by asking what it assumed

Confident-but-wrong is the #1 AI risk at work. Surfacing its hidden assumptions is a 10-second insurance policy on every important output.

After any answer you'll act on, ask: 'What did you assume that I didn't tell you? What would change your answer if it were wrong?'

1. Get the answer.
2. Ask for its assumptions + what would flip them.
3. Verify the load-bearing ones.