Library › Legal & Compliance AI › Compliance and risk

Flag gifts log entries that pass the policy thresholds

  • For Compliance and risk
  • You practice Catching mistakes and made-up facts
  • Time about 5 minutes
  • Level Intermediate

About this exercise

Review the gifts log against the two thresholds in the policy and flag entries for the compliance reviewer. Reply with a markdown table with the columns Entry | Reason | Question for reviewer, one row per flagged entry in log order, then a last line exactly: Flagged N of 8. Reviewer decides. Flag an entry if (a) it is over $100 for one event with no pre-approval, or (b) it is the entry that first takes one recipient past $250 for the year. Show the addition in the Reason. Under 140 words.

Common mistakes when you ask AI for this

  • Flagging entry 2. The AI flags the $140 tickets even though pre-approval exists. Try: Say: "Skip entries with a pre-approval reference."
  • No running total. The AI states totals without the addition. Try: Say: "Show the addition in the Reason, like 85 + 140 + 40 = $265."
  • Flagging every later entry. The AI flags entry 8 too because the total stays above $250. Try: Say: "Flag only the entry that first passes $250."
  • Trusting the AI to count words. AI is bad at counting. Ask for a total word limit and it often lands over it. Try: Cap each part instead, for example: "each line under 15 words."

What a strong prompt includes

  • Table columns
  • Show the addition
  • Both rules
  • Exact last line
  • Pre-approval exception

Practice this now ↑

Related exercises