Yes, this is another article on AI, but please do not change the channel!
As in previous articles, my goal is to educate and demythologize AI while avoiding the onslaught of positive and negative hype with random op-ed approaches. This topic addresses a critical aspect of AI that is often overlooked — the interaction between the user and the algorithm.
In short, it is the ability to ask the “right” questions in the right sequence to achieve optimal results. This sounds simple but it is not. This begs the question as to what are the problems and what are the solutions? As we begin, there are two overriding things to keep in mind.
Understanding the Structure of AI
Subject matter experts remind us that AI, on its own, lacks human context. AI does not “think” in a conventional sense; instead, it predicts the next most likely string of information based on your input. This means that vague prompts result in generic or irrelevant outputs while specific questions unlock targeted, well-informed insights.
In other words, framing questions with well-defined constraints and clear focus significantly reduces the risk of misinformation. In fact, AI researchers tell us that strategic questioning can increase AI efficiency and accuracy by up to 40%. This falls under the umbrella of what AI scientists call “prompt engineering.”
Why Prompt Engineering Matters
Prompt engineering is necessary for AI as it:
- Unlocks the potential of AI via effective questioning. This is the key to harnessing generative AI, allowing for efficient and more accurate information gathering.
- Provides AI with detailed context to understand nuanced requests, thus reducing irrelevant noise.
- Improves outcomes and leads to specific answers, better data analysis and more informed decisions.
- Shifts human role from simply receiving answers to becoming “curators and interpreters of information, requiring critical thinking to formulate insightful queries.”
- Promotes deeper understanding where users discover gaps in their own knowledge and clarify their goals.
How to Draft Stronger Prompts
Below are several tips to help you develop stronger inquiry skills and create greater prompts when using AI:
- Continuously ask “why” and “how.” Curiosity leads to deeper thinking and better questions.
- Be specific and phrase your question in a clear, concise manner.
- Employ context and background details to reduce the likelihood of hallucination.
- Define the role and who you want the AI to be (“Act as a sales management expert”).
- Specify exactly the task you want done (“Generate five catchy taglines”).
- Specify the format of the answer (“in a bulleted list”).
- Iterate and refine your questions to improve the results.
- Use open-ended questions for creativity and broad exploration.
- Use closed-ended questions for verifying facts or specific data points.
- Ask follow-up questions to dig deeper until you fully understand the topic.
Final Thoughts
Asking a question is simple but asking the right question is not. In fact, the 2500-year-old quote, “The one who knows all the answers has not been asked all the questions,” by Chinese philosopher Confucius reminds us that curiosity is as important as knowledge.
As AI continues to evolve and grow in popularity, remember that anyone, regardless of their job or skill level, can join the journey. All it takes is the courage to explore and ask the right questions!
Alan C. Brawn, CTS, DSCE, DSDE, DSNE, DCME, DSSP, ISF-C, is principal of Brawn Consulting.
