Library › Research AI › Finance and accounting

Check a chatbot's budget variance explanation against the report

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

About this exercise

You pasted your department's budget-vs-actual report into a chatbot and it explained the variances. You want one numbered line per claim, under 50 words total, labeled Matches, Wrong, or Not in report, with the report's figure after any Wrong label. Use only the report, plain text, nothing else.

Common mistakes when you ask AI for this

  • Trusting the explanation. A plausible reason, like an agency retainer, sounds right even though nothing in the report says it. Try: Add: "If the report doesn't say it, label it Not in report."
  • Missing the direction. The dollar amount is right, so a quick check passes 'over' when it was 'under'. Try: Say: "Check amount and direction (over or under)."
  • Getting new analysis. The AI explains the variances itself instead of checking the chatbot's claims. Try: Say: "Label each claim only."
  • 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 10 words."
  • Getting markdown back. AI defaults to bold labels and headings, which show up as stray symbols in your review notes. Try: Ask for "plain text, no bold, no headings."

What a strong prompt includes

  • Gives exact labels
  • Names the source
  • Checks direction
  • Sets a length limit
  • Asks for plain text

Practice this now ↑

Related exercises