Don't be the bot
Brief an AI like you would brief a capable stranger, then make it criticise your thinking instead of flattering it.
Editorial. No product being sold here.
Average in, average out
Most people type into an AI the way they type into a search box: a few words, a vague hope, then a verdict on the answer. The answer comes back smooth, competent and completely generic, and that feels like the tool's fault. It usually isn't. A thin request has no information in it about what a good answer would look like, so the model does the only sensible thing available to it and produces the most average response that fits. Being generic is the safest possible guess when nothing in the question rules anything out.
Keep & share
The takeaway
The quality of the brief decides the quality of the answer, and agreement from an AI is not evidence that you are right.
Next step: Take the last prompt you wrote, add the audience, the real material and one request for criticism, then run it again and compare.