Why It Matters
The quality of your AI output depends heavily on how you craft your prompts. Think of prompting as “programming with words” — clear, structured input leads to better results and a reduced number of calls.
UVM-Specific Tips
- Respect Data Policies: Use your UVM account (look for the Green Shield Icon to ensure Enterprise Data Protection is enabled)
- Verify Outputs: Always fact-check AI responses before sharing.
- Understand Limitations: LLMs may hallucinate; use them as assistants, not authorities.
- Prompt Gallery: Save and reuse effective prompts via Copilot’s Prompt Gallery feature for consistency.
Core Principles
Provide Context
- Explain the background, audience, and purpose.
- Example: “Write a 200-word summary of this report for a non-technical audience.”
Be Specific
- Include details like tone, format, and constraints.
- Example: “Generate five social media captions under 100 characters, using a friendly tone.”
Use Step-by-Step Instructions
- Break complex tasks into smaller steps.
- Example: “First outline the key points, then draft a summary.”
Iterate and Refine
- Start broad, then refine based on the AI’s response.
- Use follow-up prompts to clarify or adjust.
Leverage Examples
- Provide sample outputs or templates.
- Example: “Here’s an example of the tone I want: [insert example].”
Set Role or Perspective
- Assign the AI a role for better alignment.
- Example: “Act as a cybersecurity analyst and explain the risks of weak passwords.”
Common Pitfalls
- Vague prompts like “Write something about AI” → lead to generic results.
- Overloading with multiple unrelated tasks in one prompt.
- Ignoring iterative refinement.