Take the request “explain how to use video calls” and add context to it: for whom and why. For example, for whom exactly (age, level) and for what purpose. The answer is in plain text, one or two phrases.
The answer to your situationThe neural network does not know you and your situation - it only sees the words of the request. It’s like explaining the way: a visitor needs to be told in detail, but a neighbor needs only a few words. If you add who the answer is for and why, the AI will adjust both the words and the depth. Compare: “explain how to use video calls” and “explain to my 70-year-old mother, who first picked up a smartphone, how to make video calls.” In the second case, the answer will be simpler, without complex words, in large steps. Technique: at the end of the request, add one phrase “for whom” and one “why”. For example: “for a newbie to call his grandchildren.” A common mistake: people think that the AI will “guess” their level and goal. He won't guess - he will average. A few words of context saves ten minutes of rework. And don’t worry that the context is “personal”: the role and purpose are enough, without names and addresses - we will learn this separately.
The model's response will appear here...