Library › Research AI › Small business owners

Catch a chatbot's mistakes about a supplier's return policy

Practice: Catching mistakes and made-up facts. About 5 minutes.

About this exercise

You asked a chatbot how to return a damaged shipment of mugs to your supplier, and you want to check its answer against the supplier's actual return policy. You want only the errors, as numbered lines in the form "Chatbot says X; policy says Y", under 60 words total, plain text. Use only the policy, skip lines that match it, and don't rewrite the answer.

Common mistakes when you ask AI for this

  • Asking if the answer is right. The AI rewrites a corrected answer, and you can't see what the chatbot got wrong. Try: Say: "List only the errors. Don't rewrite."
  • Not naming the source. The AI may judge by typical return policies, not Clayworks' actual one. Try: Add: "The policy is the only source of truth."
  • Flagging correct lines. To look thorough, the AI lists matching steps too. Try: Say: "Skip lines that match the policy."
  • Trusting the AI to count words. AI is bad at counting. Ask for a total word limit and it often lands 10-20% over. Try: Cap each line: "each line under 14 words."
  • Getting markdown back. AI defaults to bold labels and headings, which show up as stray symbols in your notes. Try: Ask for "plain text, no bold, no headings."

What a strong prompt includes

  • Names the source of truth
  • Asks for errors only
  • Gives the line format
  • Sets a length limit
  • Asks for plain text

Practice this now ↑

Related exercises