{"id":88,"date":"2025-09-02T17:03:35","date_gmt":"2025-09-02T17:03:35","guid":{"rendered":"https:\/\/books.nbsplabs.com\/ai-lit-intro\/?post_type=chapter&#038;p=88"},"modified":"2025-10-13T13:57:42","modified_gmt":"2025-10-13T13:57:42","slug":"chapter-2-2-prompt-engineering-101","status":"publish","type":"chapter","link":"https:\/\/books.nbsplabs.com\/ai-lit-intro\/chapter\/chapter-2-2-prompt-engineering-101\/","title":{"raw":"2.2. Weak vs. Strong Prompts","rendered":"2.2. Weak vs. Strong Prompts"},"content":{"raw":"When working with AI, the difference between results that miss the mark and results that deliver exactly what you need often comes down to a few words. Weak prompts tend to produce generic or off-target responses, while well-crafted prompts are precise, context-rich, and tuned to your purpose (Sahoo et al., 2024).\r\n<h2>What Makes a Prompt Weak?<\/h2>\r\nWeak prompts often share certain characteristics:\r\n<ul>\r\n \t<li>Vagueness: Lacking enough information about audience, purpose, or form.<\/li>\r\n \t<li>No constraints: No guidance about tone, format, length, or style.<\/li>\r\n \t<li>No context: No background given or relevant details left out.<\/li>\r\n \t<li>No role or viewpoint: The prompt doesn\u2019t specify whether the AI should be \u201cexpert,\u201d \u201clayperson,\u201d \u201cteacher,\u201d etc., when responding.<\/li>\r\n<\/ul>\r\nExamples of weak prompts:\r\n<ul>\r\n \t<li><em>\u201cTell me about climate change.\u201d<\/em><\/li>\r\n \t<li><em>\u201cSummarize Shakespeare.\u201d<\/em><\/li>\r\n \t<li><em>\u201cExplain photosynthesis.\u201d<\/em><\/li>\r\n \t<li><em>\u201cWrite an email to a prospective customer.\u201d<\/em><\/li>\r\n \t<li><em>\u201cTell me a short story for kids.\u201d<\/em><\/li>\r\n<\/ul>\r\nIn each case, the AI may respond, but the answer will likely be too broad, unfocused, or unhelpful in practice.\r\n<h2>What Makes a Prompt Strong?<\/h2>\r\nA strong prompt provides the AI with clarity, context, and constraints. It tells the system not just what you want, but how you want it presented. Strong prompts consider audience, format, scope, and tone.\r\n\r\nExamples of strong prompts in different domains:\r\n<ul>\r\n \t<li><strong>Education:<\/strong> <em>\u201cExplain climate change to a group of 5th graders using a playground analogy, in fewer than 150 words.\u201d<\/em><\/li>\r\n \t<li><strong>Research:<\/strong> <em>\u201cSummarize the methodology of this abstract in one sentence, focusing only on sample size, instruments, and analysis.\u201d<\/em><\/li>\r\n \t<li><strong>Business:<\/strong> <em>\u201cWrite a professional email to a client apologizing for a delayed shipment. Keep it under 150 words, maintain a positive tone, and suggest a 10% discount as goodwill.\u201d<\/em><\/li>\r\n \t<li><strong>Creative Writing:<\/strong> <em>\u201cWrite a 200-word bedtime story for a 6-year-old about a curious cat exploring the moon. Use simple, rhythmic language.\u201d<\/em><\/li>\r\n<\/ul>\r\nNotice how these prompts provide clear direction. They guide the AI toward the response you want, without leaving it to guess.\r\n<h2>Video Example<\/h2>\r\nThis short video walks through the essentials of prompt engineering, showing how weak prompts differ from strong ones and offering tips for structuring better inputs:\r\n<div class=\"pb-embed\">\r\n\r\n[embed]https:\/\/www.youtube.com\/embed\/LAF-lACf2QY[\/embed]\r\n\r\n<\/div>\r\n<h2>Common Trade-Offs &amp; Nuances<\/h2>\r\nStrong prompts are not always better in every scenario. Sometimes you might deliberately use weak prompts for creativity or discovery. For example, vague prompts can surface surprising ideas or creative interpretations. But they also require more post-editing, carry greater risk of irrelevance or errors, and can waste time if the output is far from what you needed.\r\n\r\nAnother nuance: over-constraining a prompt can stifle creativity or miss unexpected angles. The best prompts often balance specificity with openness \u2014 enough guidance to steer, but enough freedom to allow the AI to contribute something new.\r\n<h2>Analogy (click to expand)<\/h2>\r\n<details class=\"pb-details\" style=\"border: 2px solid #2196F3;background-color: #f0f8ff;padding: 1em;margin: 1em 0;border-radius: 6px\"><summary>\ud83d\udcd6 Ordering at a Restaurant<\/summary>Imagine walking into a restaurant and saying, \u201cI'd like something good for dinner, maybe chicken.\u201d You\u2019ll likely get something edible, but it might not be what you wanted. Instead, if you say, \u201cI\u2019d like a grilled chicken sandwich with no mayo, a side salad, and water with lemon,\u201d you\u2019re much more likely to get exactly the meal you had in mind. Prompts work the same way: the more specific you are, the closer the AI gets to your intended result.\r\n\r\n<\/details>\r\n<h2>\ud83d\udcda Weekly Reflection Journal<\/h2>\r\n<div style=\"border: 2px solid #4CAF50;background-color: #f0fff0;padding: 1em;margin: 1em 0;border-radius: 6px\">\r\n\r\n<strong>Reflection Prompt<\/strong>: Think of a task you regularly perform in your teaching, research, workplace, or creative life. Write one weak prompt for that task, then rewrite it as a strong prompt with specific audience, format, and scope. What differences do you notice in clarity and usefulness?\r\n\r\n<\/div>\r\n<h2>Self-Check: Weak vs. Strong Prompts Quiz<\/h2>\r\n[h5p id=\"20\"]\r\n<h2>Looking Ahead<\/h2>\r\nNow that you\u2019ve seen the contrast between weak and strong prompts, the next chapter will introduce a toolkit of strategies for building strong prompts from the ground up. These strategies include role assignment, adding context, and refining tone and format \u2014 skills that will help you become a more intentional communicator with AI.","rendered":"<p>When working with AI, the difference between results that miss the mark and results that deliver exactly what you need often comes down to a few words. Weak prompts tend to produce generic or off-target responses, while well-crafted prompts are precise, context-rich, and tuned to your purpose (Sahoo et al., 2024).<\/p>\n<h2>What Makes a Prompt Weak?<\/h2>\n<p>Weak prompts often share certain characteristics:<\/p>\n<ul>\n<li>Vagueness: Lacking enough information about audience, purpose, or form.<\/li>\n<li>No constraints: No guidance about tone, format, length, or style.<\/li>\n<li>No context: No background given or relevant details left out.<\/li>\n<li>No role or viewpoint: The prompt doesn\u2019t specify whether the AI should be \u201cexpert,\u201d \u201clayperson,\u201d \u201cteacher,\u201d etc., when responding.<\/li>\n<\/ul>\n<p>Examples of weak prompts:<\/p>\n<ul>\n<li><em>\u201cTell me about climate change.\u201d<\/em><\/li>\n<li><em>\u201cSummarize Shakespeare.\u201d<\/em><\/li>\n<li><em>\u201cExplain photosynthesis.\u201d<\/em><\/li>\n<li><em>\u201cWrite an email to a prospective customer.\u201d<\/em><\/li>\n<li><em>\u201cTell me a short story for kids.\u201d<\/em><\/li>\n<\/ul>\n<p>In each case, the AI may respond, but the answer will likely be too broad, unfocused, or unhelpful in practice.<\/p>\n<h2>What Makes a Prompt Strong?<\/h2>\n<p>A strong prompt provides the AI with clarity, context, and constraints. It tells the system not just what you want, but how you want it presented. Strong prompts consider audience, format, scope, and tone.<\/p>\n<p>Examples of strong prompts in different domains:<\/p>\n<ul>\n<li><strong>Education:<\/strong> <em>\u201cExplain climate change to a group of 5th graders using a playground analogy, in fewer than 150 words.\u201d<\/em><\/li>\n<li><strong>Research:<\/strong> <em>\u201cSummarize the methodology of this abstract in one sentence, focusing only on sample size, instruments, and analysis.\u201d<\/em><\/li>\n<li><strong>Business:<\/strong> <em>\u201cWrite a professional email to a client apologizing for a delayed shipment. Keep it under 150 words, maintain a positive tone, and suggest a 10% discount as goodwill.\u201d<\/em><\/li>\n<li><strong>Creative Writing:<\/strong> <em>\u201cWrite a 200-word bedtime story for a 6-year-old about a curious cat exploring the moon. Use simple, rhythmic language.\u201d<\/em><\/li>\n<\/ul>\n<p>Notice how these prompts provide clear direction. They guide the AI toward the response you want, without leaving it to guess.<\/p>\n<h2>Video Example<\/h2>\n<p>This short video walks through the essentials of prompt engineering, showing how weak prompts differ from strong ones and offering tips for structuring better inputs:<\/p>\n<div class=\"pb-embed\">\n<p><iframe loading=\"lazy\" id=\"oembed-1\" title=\"Prompt engineering essentials: Getting better results from LLMs | Tutorial\" width=\"500\" height=\"281\" src=\"https:\/\/www.youtube.com\/embed\/LAF-lACf2QY?feature=oembed&#38;rel=0\" frameborder=\"0\" allowfullscreen=\"allowfullscreen\"><\/iframe><\/p>\n<\/div>\n<h2>Common Trade-Offs &amp; Nuances<\/h2>\n<p>Strong prompts are not always better in every scenario. Sometimes you might deliberately use weak prompts for creativity or discovery. For example, vague prompts can surface surprising ideas or creative interpretations. But they also require more post-editing, carry greater risk of irrelevance or errors, and can waste time if the output is far from what you needed.<\/p>\n<p>Another nuance: over-constraining a prompt can stifle creativity or miss unexpected angles. The best prompts often balance specificity with openness \u2014 enough guidance to steer, but enough freedom to allow the AI to contribute something new.<\/p>\n<h2>Analogy (click to expand)<\/h2>\n<details class=\"pb-details\" style=\"border: 2px solid #2196F3;background-color: #f0f8ff;padding: 1em;margin: 1em 0;border-radius: 6px\">\n<summary>\ud83d\udcd6 Ordering at a Restaurant<\/summary>\n<p>Imagine walking into a restaurant and saying, \u201cI&#8217;d like something good for dinner, maybe chicken.\u201d You\u2019ll likely get something edible, but it might not be what you wanted. Instead, if you say, \u201cI\u2019d like a grilled chicken sandwich with no mayo, a side salad, and water with lemon,\u201d you\u2019re much more likely to get exactly the meal you had in mind. Prompts work the same way: the more specific you are, the closer the AI gets to your intended result.<\/p>\n<\/details>\n<h2>\ud83d\udcda Weekly Reflection Journal<\/h2>\n<div style=\"border: 2px solid #4CAF50;background-color: #f0fff0;padding: 1em;margin: 1em 0;border-radius: 6px\">\n<p><strong>Reflection Prompt<\/strong>: Think of a task you regularly perform in your teaching, research, workplace, or creative life. Write one weak prompt for that task, then rewrite it as a strong prompt with specific audience, format, and scope. What differences do you notice in clarity and usefulness?<\/p>\n<\/div>\n<h2>Self-Check: Weak vs. Strong Prompts Quiz<\/h2>\n<div id=\"h5p-20\">\n<div class=\"h5p-iframe-wrapper\"><iframe id=\"h5p-iframe-20\" class=\"h5p-iframe\" data-content-id=\"20\" style=\"height:1px\" src=\"about:blank\" frameBorder=\"0\" scrolling=\"no\" title=\"2.2. Weak vs. Strong Prompts Quiz\"><\/iframe><\/div>\n<\/div>\n<h2>Looking Ahead<\/h2>\n<p>Now that you\u2019ve seen the contrast between weak and strong prompts, the next chapter will introduce a toolkit of strategies for building strong prompts from the ground up. These strategies include role assignment, adding context, and refining tone and format \u2014 skills that will help you become a more intentional communicator with AI.<\/p>\n","protected":false},"author":6,"menu_order":2,"template":"","meta":{"pb_show_title":"on","pb_short_title":"","pb_subtitle":"","pb_authors":["avas"],"pb_section_license":"cc-by-nc-sa","_links_to":"","_links_to_target":""},"chapter-type":[49],"contributor":[63],"license":[57],"class_list":["post-88","chapter","type-chapter","status-publish","hentry","chapter-type-numberless","contributor-avas","license-cc-by-nc-sa"],"part":30,"jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/books.nbsplabs.com\/ai-lit-intro\/wp-json\/pressbooks\/v2\/chapters\/88","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/books.nbsplabs.com\/ai-lit-intro\/wp-json\/pressbooks\/v2\/chapters"}],"about":[{"href":"https:\/\/books.nbsplabs.com\/ai-lit-intro\/wp-json\/wp\/v2\/types\/chapter"}],"author":[{"embeddable":true,"href":"https:\/\/books.nbsplabs.com\/ai-lit-intro\/wp-json\/wp\/v2\/users\/6"}],"version-history":[{"count":14,"href":"https:\/\/books.nbsplabs.com\/ai-lit-intro\/wp-json\/pressbooks\/v2\/chapters\/88\/revisions"}],"predecessor-version":[{"id":770,"href":"https:\/\/books.nbsplabs.com\/ai-lit-intro\/wp-json\/pressbooks\/v2\/chapters\/88\/revisions\/770"}],"part":[{"href":"https:\/\/books.nbsplabs.com\/ai-lit-intro\/wp-json\/pressbooks\/v2\/parts\/30"}],"metadata":[{"href":"https:\/\/books.nbsplabs.com\/ai-lit-intro\/wp-json\/pressbooks\/v2\/chapters\/88\/metadata\/"}],"wp:attachment":[{"href":"https:\/\/books.nbsplabs.com\/ai-lit-intro\/wp-json\/wp\/v2\/media?parent=88"}],"wp:term":[{"taxonomy":"chapter-type","embeddable":true,"href":"https:\/\/books.nbsplabs.com\/ai-lit-intro\/wp-json\/pressbooks\/v2\/chapter-type?post=88"},{"taxonomy":"contributor","embeddable":true,"href":"https:\/\/books.nbsplabs.com\/ai-lit-intro\/wp-json\/wp\/v2\/contributor?post=88"},{"taxonomy":"license","embeddable":true,"href":"https:\/\/books.nbsplabs.com\/ai-lit-intro\/wp-json\/wp\/v2\/license?post=88"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}