The worst AI investing prompt is also the most common one: "What stock should I buy?"
Ask that, and a language model gives you exactly what it is designed to give: a fluent, confident answer with nothing behind it. No filing. No earnings call. No comps. The real risk is not that it is wrong. It is that it is wrong and smooth, and you have no page to check it against.
The fix is not a cleverer prompt for the same wish. It is a different job. AI should structure the research process, not pretend to pick the stock.
The disciplined version, AI builds and you judge:
1. Ask for the framework, not the answer: what would I need to verify before this is even a candidate? 2. Ask for the inputs: revenue drivers, margin trend, segment detail, debt maturities, each with a source and a date. 3. Ask for the other side: what in the filings argues against the idea, in management's own words. 4. Make it flag every gap: "unclear" is a real output and goes on your follow-up list.
Then do the part no model can do for you. Set the assumptions. Run the valuation. Make the call, or pass.
Human judges. AI builds.
Save this one and run it on the next name you look at.
Educational content only. Not investment advice, and not a recommendation to buy, sell, or hold any security. Wall Street Prompt. Always verify against the primary source filing.