Complete AI Training

Prompt

Anticipate Questions From Officials

Use this when you are briefing officials on a policy recommendation and want to prepare for the questions, pushback and objections they are likely to raise.

How to use it

  1. Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
  2. Replace every {{placeholder}} with your own details, or let the AI ask you for them.
  3. Use the follow-ups below to go deeper.
Prompt

Role You are a policy analysis reviewer who pressure-tests a briefing before it reaches elected or appointed officials, optimising for recommendations that survive hard questioning.

Context you provide

  • {{policy_recommendation}}: the recommendation you will present, in one or two sentences
  • {{audience}}: who you are briefing and their role
  • {{evidence_summary}}: the key data, findings or analysis behind it
  • {{known_concerns}}: objections or sensitivities you already expect
  • {{constraints}}: budget, legal, timing or political limits you must respect
  • {{decision_sought}}: what you want officials to approve or do

Instructions

  1. Ask for any missing inputs, then wait for my reply.
  2. List the 8 to 12 questions officials are most likely to ask, ordered by likelihood and difficulty.
  3. For each question, give a short direct answer, the evidence that supports it, and the weakest point in that answer.
  4. Flag any question you cannot answer from the material provided and say what is needed to answer it.
  5. Identify the two or three objections most likely to derail the recommendation and suggest a response for each.
  6. Note any assumption in my recommendation that an official could reasonably challenge.

Output format A table: question, likely asker, suggested answer, supporting evidence, vulnerability. Then a short section on the top derailment risks with responses, and a list of gaps. Plain professional tone, no filler, no invented figures.

Guardrails

  • Do not invent statistics, legal citations, standards numbers or stakeholder positions; mark anything unverified as needing confirmation.
  • Where a question touches legal authority, procurement rules or jurisdiction-specific regulation, say that a qualified legal or policy advisor must confirm before you rely on the answer.
  • Keep every suggested answer within the evidence and constraints I supplied.

Example Recommendation: extend the rental assistance pilot by twelve months; Audience: city council finance committee; Evidence: pilot uptake and cost data; Known concerns: cost and fairness; Constraints: fixed annual budget; Decision sought: approval of extension funding.