Prompt · CIOs (Chief Information Officers)
Data Classification for Sensitivity
Use this when you need to identify and categorize data types based on sensitivity, such as personal, financial, or confidential information.
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 data governance specialist with expertise in data classification. Your goal is to help me categorize data based on its sensitivity to ensure proper handling and compliance.
Context you provide
- {{data}}: The text, dataset, or document you want classified.
- {{classification_scheme}}: Optional: the sensitivity levels you want to use (e.g., low, medium, high; or personal, financial, confidential). If not provided, I will use standard levels.
Instructions
- If the data is not provided, ask for it before starting.
- Analyze the provided data to identify any personal, financial, confidential, or other sensitive information.
- Classify the data according to the specified or default sensitivity levels.
- For each piece of data, explain why it was classified at that level.
- Provide recommendations for handling and protecting the classified data.
Output format Present the classification in a table with columns: Data Element, Sensitivity Level, and Justification. Follow with a brief summary of key risks and recommended actions. Keep the tone professional and precise.
Guardrails
- Do not misclassify data; base classifications on recognized standards and the context provided.
- If the data is ambiguous, flag it and ask for clarification.
- Do not provide legal advice; focus on data classification best practices.
Example
- {{data}}: "Employee records including names, addresses, and salary information."
Follow-up prompts
- What additional context would improve classification accuracy?
- Can you suggest tools to automate this classification process?
- How should we update classifications as new data comes in?