AI Literacy

How Do You Write a Prompt?

Prompt engineering is important for getting the best results from AI. Clear and well-written prompts help AI provide precise answers, while unclear prompts can cause confusion.

The CLEAR Framework

The CLEAR Framework is a guide to help you write better prompts by focusing on five key principles: Concise, Logical, Explicit, Adaptive, and Reflective.

Using these principles will help you get more accurate and helpful answers from AI.

1. Concise: Keep It Short and Clear

Good prompts are short and easy to understand. By removing unnecessary details, you help the AI focus on what's most important, which leads to better-quality answers.

• Example: Instead of saying, "Can you describe the process of cell division in detail?" say, "Explain the main stages of cell division and their purpose."
• Example: Instead of saying, "Can you provide an in-depth explanation of how a computer works?" say, "Explain the basic components of a computer and their functions."

Keeping prompts concise ensures that every word matters and helps the AI respond quickly and effectively.

2. Logical: Make It Make Sense

A logical prompt is organized in a way that makes sense. It helps guide the AI step-by-step, which makes the answers clearer and more connected.

• Example: "List the steps in the scientific method, from hypothesis to conclusion."
• Example: "Explain the sequence of events in the water cycle, starting from evaporation to precipitation."

Logical prompts help the AI understand the flow of information and reduce the need for extra clarifications, making the interaction smoother.

3. Explicit: Be Specific About What You Want

Explicit prompts tell the AI exactly what you want. Clear instructions on the type, format, or details of the information you're asking for make it more likely you'll get a helpful answer.

• Example: Instead of saying, "Tell me about climate change," say, "Provide an overview of climate change, focusing on greenhouse gas emissions, rising temperatures, and potential solutions."
• Example: Instead of saying, "Tell me about World War II," say, "Summarize the main causes of World War II, including political and economic factors."

Being explicit helps the AI understand what you expect, leading to more accurate results.

4. Adaptive: Be Flexible and Adjust

Your first prompt might not give you the best answer. To get better results, you may need to adjust your prompts. Make your prompt more specific if the answer is too broad or off-topic.

• Example: If "Describe renewable energy" is too general, try "Explain how solar power works and its benefits over fossil fuels."
• Example: If "Tell me about space exploration" is too broad, try "Summarize the key milestones in space exploration from the 1960s to today."

Experiment with different ways of asking questions to get the best possible answers.

5. Reflective: Learn and Improve

Reflective prompts involve reviewing your responses and considering how to improve your prompts. Did the AI answer your question well? If not, think about what you could change.

• Example: After asking for ways to improve concentration, evaluate if the response provided practical techniques. If not, adjust your next prompt to focus on a specific technique, such as meditation, time blocking, or reducing distractions.
• Example: If you asked about learning new skills and received a vague answer, change your prompt to focus on a particular skill, like learning a new language or improving public speaking.

Reflecting on what works and what doesn't help you get better at writing prompts over time.

The Main Takeaway

Writing good prompts is a skill that takes practice and a willingness to adapt.

Use the CLEAR Framework—Concise, Logical, Explicit, Adaptive, and Reflective—to guide your prompts and improve your interactions with AI.

Clear prompts lead to better answers, which help you learn more effectively. By mastering these skills, you can make AI a helpful partner in reaching your goals.

Source: The CLEAR path: A framework for enhancing information literacy through prompt engineering