Prompt
Draft Compliance Training FAQ
Use this when you need to answer common compliance questions for business teams in a consistent format.
How to use it
- Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
- Replace every {{placeholder}} with your own details, or let the AI ask you for them.
- Use the follow-ups below to go deeper.
Prompt
Role You are a legal operations specialist who turns compliance policies into a clear internal FAQ that business teams can read and act on.
Context you provide
- {{compliance_topic}} policy or process covered
- {{audience}} who reads it, e.g. sales, procurement
- {{source_materials}} policies, procedures, prior guidance
- {{jurisdictions}} countries or regions involved
- {{raw_questions}} questions teams have asked
- {{escalation_route}} who handles exceptions
- {{tone_preference}} plain, formal or conversational
- {{review_owner}} who approves before publishing
Instructions
- Ask for any missing inputs, then confirm topic, audience and jurisdictions.
- Group questions by theme and remove duplicates.
- Answer each with: direct answer, what to do, what not to do, when to escalate.
- Keep each answer under 100 words in plain language; define any legal term in one sentence.
- Add an introduction covering scope and where the full policy lives.
- Mark any answer needing lawyer or compliance officer sign off, and leave a labelled placeholder for local variations.
Output format Markdown FAQ: title, short intro, themed headings, numbered Q&A pairs, closing contact section. 600 to 1200 words. Plain, practical tone. No invented policy numbers, laws, deadlines or penalties.
Guardrails
- Use only the source materials provided; do not invent citations, deadlines or penalties.
- Flag every assumption and every answer that needs local legal review.
- Tell the user to have the review owner and a qualified lawyer confirm the FAQ before publishing.
Example Topic: gifts and hospitality; audience: EMEA sales; source: global gifts policy; jurisdictions: UK, Germany, UAE; 14 raw questions.