How to stop ChatGPT from BS

When using ChatGPT, one of the most important yet often overlooked steps is to explicitly set the rules for the AI before asking your main questions. By doing this, you essentially reduce the chance of the model making assumptions or 'guessing' information that it doesn't actually have. For example, starting your prompt with instructions like "Only provide answers based on verified facts" or "Avoid speculation and stick to known data" can significantly improve the reliability of the responses. From my experience, another useful approach is to provide context or background data upfront within the prompt. This helps ChatGPT understand the scope and boundaries of the conversation, reducing vague or fabricated details. Instead of vague queries, detailed prompts with concrete examples or constraints tend to guide the AI more effectively. Lastly, iterative prompting can help refine answers. If the first response feels uncertain or inaccurate, follow up with clarifying questions or ask the model to explain its reasoning. This often surfaces where assumptions were made and gives you a chance to steer the conversation toward precise information. Overall, stopping AI guessing involves a combination of setting explicit rules, giving sufficient context, and actively engaging with the answers received. These methods align well with the popular advice to 'STOP AI GUESSING Set the Rules First,' which ensures that ChatGPT's outputs are trustworthy and useful for creators and anyone relying on AI-generated content.