Practical prompting, backed by research

One phrase, "consider diversity," and the model starts inventing a patient's race that wasn't in the question. This breakdown of Demographic Injection shows when an equity instruction makes AI mix in nonexistent details, and how to block it with one line in your prompt. Without the phrase, the model fabricates demographics in 0.7% of answers; with the DEI phrase, in 33%, a 47-fold jump. The model confuses general statistics about a category of people with a fact about the specific person in the question, and in 99.8% of cases where the fabrication changes the answer, the answer becomes wrong. The effect held across all 47 models tested, from GPT-5 to small open ones.

Discovery: ChatGPT picks its advice based on the tone of a message, not the actual danger level. "I feel sick, ate something bad" — instant advice about bananas and toast. "Chest pain" — instant alarm and questions about the heart. The entry-to-care prompt method forces the model to first ask about critical facts — age, chronic conditions, duration — before giving any advice. One system prompt flips the order: questions first, diagnosis later. In one test, the model missed diabetic ketoacidosis entirely — it simply forgot to ask about insulin in time.