--- title: I stripped my prompts down after OpenAI's new guide, and... description: OpenAI's new prompting guidance challenges years of conventional wisdom that said more context and more instructions always produce better results. The company... canonical: https://daily.dev/posts/i-stripped-my-prompts-down-after-openai-s-new-guide-and-my-results-got-better-y9gjbkpk9 twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: I stripped my prompts down after OpenAI's new guide, and my results got better | daily.dev og:description: OpenAI's new prompting guidance challenges years of conventional wisdom that said more context and more instructions always produce better results. The company... og:url: https://daily.dev/posts/i-stripped-my-prompts-down-after-openai-s-new-guide-and-my-results-got-better-y9gjbkpk9 og:image: https://api.daily.dev/og/posts/y9gJbKPk9.png og:image:alt: I stripped my prompts down after OpenAI's new guide, and my results got better og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # I stripped my prompts down after OpenAI's new guide, and my results got better **[XDA Developers](https://daily.dev/sources/xda-developers)** · 4 min read · 1 upvotes · 0 comments ## Summary OpenAI's new prompting guidance challenges years of conventional wisdom that said more context and more instructions always produce better results. The company now recommends leaner prompts — stating each instruction once, removing redundant rules and examples, and avoiding repeated approval-seeking phrases that make models overly cautious. Internal evaluations showed leaner system prompts improved coding-agent scores by 10–15% while reducing token usage and cost. The guidance also notes that newer models like GPT-5.6 are more concise by default, so vague instructions like 'be concise' can backfire. Instead, specify what a shorter response must still preserve. The broader shift reflects that modern models are better at inferring intent, making heavy scaffolding less necessary and sometimes counterproductive. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### Does OpenAI recommend adding more detail and examples to prompts or stripping them down? OpenAI's newer guidance recommends stripping prompts down rather than piling on more context, rules, and examples. It advises stating each instruction once and removing repeated instructions or examples that aren't pulling their weight. In OpenAI's own internal coding-agent evaluations, leaner system prompts improved scores by roughly 10 to 15 percent while substantially reducing token usage and cost. _Developers refining prompt strategy for coding agents can find similar practical AI workflow guidance on daily.dev._ ### Why doesn't telling GPT-5.6 to 'be concise' work as well as it used to? GPT-5.6 is already more concise by default than GPT-5.5, so vague brevity instructions like 'be concise' or 'keep it short' can push the model too far and produce a shorter answer than intended. OpenAI recommends instead specifying what a shorter response must still preserve, such as the conclusion, supporting evidence, caveats, and next steps, while naming the lower-value material to cut. _Anyone tuning response length in GPT-5.6 prompts can track these model-behavior shifts on daily.dev._ ### Why does repeating 'ask first' or 'wait for approval' throughout a prompt make an AI model overly cautious? Scattering repeated approval instructions like 'ask first' or 'don't make changes' throughout a prompt can make GPT-5.6 overly cautious, causing it to ask for confirmation even on safe actions it was already meant to perform. OpenAI recommends instead defining clear autonomy and approval boundaries once, since GPT-5.6 is designed to be more proactive on multi-step tasks. _daily.dev helps developers keep pace with prompting behavior changes when building agentic AI workflows._ ## Similar posts on daily.dev - [Your prompts are limiting your AI more than your model choice ever could](https://daily.dev/posts/your-prompts-are-limiting-your-ai-more-than-your-model-choice-ever-could-off0guzyq) · XDA Developers · 1 upvotes · 0 comments - [Implementing Prompt Compression to Reduce Agentic Loop Costs](https://daily.dev/posts/implementing-prompt-compression-to-reduce-agentic-loop-costs-tz6xlf8uj) · Machine Learning Mastery · 27 upvotes · 1 comments - [20 approaches to writing better AI prompts](https://daily.dev/posts/20-approaches-to-writing-better-ai-prompts-tlrg0rbbe) · InfoWorld · 0 upvotes · 0 comments --- Tags: [#llm](https://daily.dev/tags/llm), [#openai](https://daily.dev/tags/openai), [#chatgpt](https://daily.dev/tags/chatgpt), [#prompt-engineering](https://daily.dev/tags/prompt-engineering) [View this post on daily.dev](https://daily.dev/posts/i-stripped-my-prompts-down-after-openai-s-new-guide-and-my-results-got-better-y9gjbkpk9) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"I stripped my prompts down after OpenAI's new guide, and my results got better","url":"https://daily.dev/posts/i-stripped-my-prompts-down-after-openai-s-new-guide-and-my-results-got-better-y9gjbkpk9","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/i-stripped-my-prompts-down-after-openai-s-new-guide-and-my-results-got-better-y9gjbkpk9"},"datePublished":"2026-08-10T14:32:19.879Z","dateModified":"2026-09-13T21:09:10.824Z","description":"OpenAI's new prompting guidance challenges years of conventional wisdom that said more context and more instructions always produce better results. The company...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/ee1aa763aa3b53dad3868b8721808cbf?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/ee1aa763aa3b53dad3868b8721808cbf?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"XDA Developers","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"XDA Developers","logo":"https://media.daily.dev/image/upload/s--kCg6yyAP--/f_auto,q_auto/v1774964407/logos/xda-developers?_a=BAMAMiWQ0","url":"https://daily.dev/sources/xda-developers"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/i-stripped-my-prompts-down-after-openai-s-new-guide-and-my-results-got-better-y9gjbkpk9","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"llm,openai,chatgpt,prompt-engineering","timeRequired":"PT4M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"XDA Developers","item":"https://daily.dev/sources/xda-developers"},{"@type":"ListItem","position":3,"name":"I stripped my prompts down after OpenAI's new guide, and my results got better"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/i-stripped-my-prompts-down-after-openai-s-new-guide-and-my-results-got-better-y9gjbkpk9#faq","mainEntity":[{"@type":"Question","name":"Does OpenAI recommend adding more detail and examples to prompts or stripping them down?","acceptedAnswer":{"@type":"Answer","text":"OpenAI's newer guidance recommends stripping prompts down rather than piling on more context, rules, and examples. It advises stating each instruction once and removing repeated instructions or examples that aren't pulling their weight. In OpenAI's own internal coding-agent evaluations, leaner system prompts improved scores by roughly 10 to 15 percent while substantially reducing token usage and cost. Developers refining prompt strategy for coding agents can find similar practical AI workflow guidance on daily.dev."}},{"@type":"Question","name":"Why doesn't telling GPT-5.6 to 'be concise' work as well as it used to?","acceptedAnswer":{"@type":"Answer","text":"GPT-5.6 is already more concise by default than GPT-5.5, so vague brevity instructions like 'be concise' or 'keep it short' can push the model too far and produce a shorter answer than intended. OpenAI recommends instead specifying what a shorter response must still preserve, such as the conclusion, supporting evidence, caveats, and next steps, while naming the lower-value material to cut. Anyone tuning response length in GPT-5.6 prompts can track these model-behavior shifts on daily.dev."}},{"@type":"Question","name":"Why does repeating 'ask first' or 'wait for approval' throughout a prompt make an AI model overly cautious?","acceptedAnswer":{"@type":"Answer","text":"Scattering repeated approval instructions like 'ask first' or 'don't make changes' throughout a prompt can make GPT-5.6 overly cautious, causing it to ask for confirmation even on safe actions it was already meant to perform. OpenAI recommends instead defining clear autonomy and approval boundaries once, since GPT-5.6 is designed to be more proactive on multi-step tasks. daily.dev helps developers keep pace with prompting behavior changes when building agentic AI workflows."}}]} ```