Library › Leadership AI › Executives

Rate diligence findings with a stated rule and questions for counsel

  • For Executives
  • You practice Setting limits and rules
  • Time about 5 minutes
  • Level Advanced

About this exercise

Rate each diligence finding using this rule: High if it affects more than 10 percent of revenue or involves a lawsuit; Medium if a required document is missing; otherwise Low. Build a table with the header Finding | Severity | Question for counsel, one row per finding, with each question starting "Can counsel" and ending with a question mark. Do not say whether anything is legal, illegal or enforceable. Plain text, under 190 words.

Common mistakes when you ask AI for this

  • Following the banker's opinion. The AI rates the GPL item High. Try: Say: "Apply the rule exactly; ignore opinions in the notes."
  • Giving legal conclusions. The AI says the claim is weak or the IP is invalid. Try: Say: "Do not say whether anything is legal or enforceable; ask counsel."
  • Statements instead of questions. The AI writes recommendations. Try: Say: "Each question starts with Can counsel and ends with ?."
  • Wrong table. The AI adds a rationale column or skips rows. Try: Say: "Exactly this header and five rows."
  • Markdown. The AI bolds severities. Try: Say: "Plain text."
  • 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

  • Gives the header
  • States the rating rule
  • Says to apply the rule only
  • Sets question form
  • Bans legal conclusions

Practice this now ↑

Related exercises