AI Prompting

Welcome to the episode. Today we’re talking about AI prompting: how to give an AI tool instructions that lead to more useful, relevant answers. You don’t need secret phrases or technical expertise to get started. Often, the difference between a vague response and a helpful one comes down to how clearly you explain what you need.

First, give the AI some context. A prompt like “Write about gardening” leaves a lot open: Who is the audience? What kind of gardening? How long should the response be? Try adding the situation instead: “Write a friendly, 200-word introduction to growing herbs on an apartment windowsill for beginners.” The extra details help the AI understand the task and shape its response around your goal.

Second, be specific about the result you want. Tell the AI what format, tone, and level of detail to use. You might ask for a short email, a list of ideas, or a step-by-step explanation in plain language. You can also describe what to avoid, such as jargon, an overly formal tone, or unsupported claims. These directions don’t guarantee a perfect answer, but they give the tool a clearer target.

Third, treat prompting as a conversation, not a one-shot test. If the first answer is close but not quite right, explain what needs to change: “Make it warmer,” “Cut the introduction,” or “Add an example for someone on a tight budget.” You can also ask the AI to revise one section rather than start over. Good AI prompting often involves a few rounds of refinement, just as writing or brainstorming with another person might.

It can help to share an example, too. If you have a style or structure in mind, include a short sample and say what you want the AI to follow—perhaps the sentence length, level of formality, or organization. Be clear about which features matter. An example is a guide, not a guarantee, and the result should still be checked for accuracy and fit.

That brings us to an important point: AI-generated answers can be mistaken, incomplete, or out of date. For anything important—especially health, legal, financial, or factual information—verify the details with reliable sources or a qualified professional. Avoid entering private or sensitive information unless you understand how the tool handles it. A well-written response can sound confident even when it needs correction, so human judgment remains essential.

Here’s a simple prompt pattern to try: describe the task, give the context, specify the format and tone, and explain any limits. For example: “Suggest five low-cost vegetarian lunches for a busy student. Keep each idea to two sentences, use everyday ingredients, and include one option that can be prepared ahead.” Then review the answer and ask for changes where needed.

The takeaway is simple: clear prompts make it easier to get useful results, and thoughtful follow-up makes them better. Start with your goal, provide the details that matter, and check the response before relying on it. That’s the practical side of AI prompting—and a good way to make these tools work more effectively for you.