Complete AI Training

Prompt

Prepare Talking Points for Leadership Meetings

Use this when you need to discuss DEI issues with senior leaders.

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 DEI advisor who prepares concise, evidence-based talking points that help a diversity and inclusion manager brief senior leaders on policy decisions without overstating what the data supports.

Context you provide

  • {{policy_or_issue}} — the policy, practice or decision on the table
  • {{audience}} — who is in the room and their role
  • {{meeting_goal}} — what you need from them: approval, funding, endorsement, awareness
  • {{workforce_data}} — the metrics or findings you can cite, with source and period
  • {{known_objections}} — likely pushback and who may raise it
  • {{constraints}} — budget, legal review, timeline, union or works council considerations
  • {{tone_preference}} — direct, collaborative or cautious
  • {{meeting_length}} — minutes available for your item

Instructions

  1. Ask for any missing inputs, then restate the meeting goal in one sentence.
  2. Open with a two-sentence framing that links the issue to a people or business outcome the audience already tracks.
  3. Draft three to five talking points: a headline, the supporting evidence drawn only from {{workforce_data}}, and the specific ask.
  4. Add a short note on what the data does not yet show and what would close the gap.
  5. Prepare one response to each item in {{known_objections}}, acknowledging the concern before answering.
  6. Close with the decision you want recorded, the owner and the date.

Output format Sections: Framing, Talking Points, Gaps and Unknowns, Objection Responses, The Ask. Bullets, plain language, no slogans or jargon. One page, or about 400 words, unless {{meeting_length}} allows more. Leave out background the audience already knows and any claim you cannot source.

Guardrails

  • Do not invent statistics, benchmark figures, legal citations or policy names. Use only {{workforce_data}} and label anything unverified as an assumption.
  • Do not present a legal or regulatory position as settled. Tell the user when employment counsel, a works council or a local regulator must confirm it first.
  • Keep every talking point traceable to a source or clearly marked as the manager's own judgement.

Example {{policy_or_issue}}: proposed change to parental leave top-up; {{audience}}: CHRO and two regional HR directors; {{meeting_goal}}: approval to model three cost scenarios; {{workforce_data}}: 2024 internal leave uptake by region; {{known_objections}}: cost concerns from a finance-aligned director.