{
  "id": 16395162,
  "title": "The Aircraft Manual Prompt That Cuts AI Waffle: ASD-STE100 for ChatGPT and Claude",
  "url": "https://urgent.news/2026/10/25/the-aircraft-manual-prompt-that-cuts-ai-waffle-asd-ste100-for-chatgpt",
  "topic": "ai",
  "section": "AI",
  "published": "2026-10-25T00:13:58.000Z",
  "source": {
    "name": "Dev.to",
    "slug": "dev-to",
    "url": "https://dev.to/blobxiaoyao/the-aircraft-manual-prompt-that-cuts-ai-waffle-asd-ste100-for-chatgpt-and-claude-39k4"
  },
  "original_language": "en",
  "account": "Chatbot responses often feel bloated and confusing, with the first answer repeating the question, the second hedging, the third delivering useful information after a convoluted structure, and the final response summarizing. The usual solution is to request brevity or simplicity, which typically works for roughly half the replies. However, a more effective approach comes from an unexpected source: the aircraft maintenance manual.\n\nIn a post on October 2, 2026, former OpenAI head and Tesla engineer Andrej Karpathy shared insights on obtaining clearer AI model outputs. Recognizing that AI models generate text and code faster than humans can process them, Karpathy proposed ranking AI output formats based on human digestibility. The hierarchy ranges from controlled plain text to custom explainer videos.\n\nThe key to improving step one, the plain text response, lies in instructing the AI to write in a standardized format called Simplified Technical English (STE). This standard, also known as ASD-STE100, originated in the European aerospace industry in 1986. Its purpose is to minimize ambiguity in technical documents, as even minor misunderstandings could have serious consequences in aircraft maintenance.\n\nThree main rules of STE carry the most weight. First, sentences should express one idea only, limiting them to around 20 words for procedural instructions and 25 words for descriptive content. This forces the model to be concise and directly convey a single fact or command. Second, the model should use active voice with a clear subject performing an action. Passive constructions, where the subject is hidden, make it harder to understand who is doing what. Lastly, each approved word in STE has a single meaning and part of speech. This eliminates ambiguity caused by elegant variation in language usage.\n\nKarpathy believes that explicitly requesting text written in the STE format is more effective than simply asking for concise or simplified language. This is because STE is a documented standard with a clear set of rules, while \"be concise\" is an open-ended directive that leaves interpretation to the model. By providing a specific style guide, the model has a fixed framework to follow, reducing variability in its responses.\n\nThe broader lesson from this tip extends beyond AI interaction. When seeking a particular style or format for writing, it is often more effective to look for an existing standard that defines it, rather than attempting to articulate the desired characteristics yourself. Style guides for plain language, newspapers, legal documents, or scientific reports all serve the same purpose: to establish a consistent, clear, and easily understandable approach to writing.",
  "summary": "You ask a chatbot a simple question and get four paragraphs back. The first one repeats your question. The second one hedges. The third one finally says something useful, buried inside a sentence with two subordinate clauses and a \"however\". The fourth one summarizes. The usual fix is to type \"be concise\" or \"explain it simply\". Both work for about one reply. A better fix comes from a place that…",
  "key_points": [
    "Chatbot responses often feel bloated and confusing.",
    "Andrej Karpathy proposed using Simplified Technical English (STE) for clearer AI outputs.",
    "STE rules force concise, active voice, and single meaning words to reduce variability."
  ],
  "editors_take": "Applying rigid aerospace documentation standards to large language models replaces vague instructions with a fixed framework that eliminates linguistic ambiguity and forces more digestible, direct communication.",
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}