A user found that adding one sentence to nearly every ChatGPT prompt improved results. The sentence is: "Before answering, ask me any clarifying questions you need. If you already have enough information, answer immediately and explain any assumptions you're making." The article notes that when ChatGPT gives a disappointing answer, it is usually because it guessed what the user meant instead of asking what they actually wanted. This issue persists across every ChatGPT model used. The added sentence does not make ChatGPT smarter but makes it less likely to head off in the wrong direction, and it works in Claude, Gemini and most other AI chatbots tried. The result is that the user spends far less time correcting responses after the fact. The article explains that chatbots or large language models (LLMS) are designed to predict the most likely next words based on information provided, and when key details are missing, they simply make their best guess.