AI's value is not one-shotting, humans one-shot better. Its value is iterating from 80% to 99% far faster than you could. So: give it one real, small task, then push back twice. Feel the loop. And if an answer loses you, don't nod along. Say 'Wait, I don't follow. Re-explain that in plain words, from the start.' Asking it to re-pitch costs one message; building on something you didn't understand costs much more.
Where these ideas come from: The wait-what skill (open-source skills library (MIT)). Paraphrased here.
Do this
0/6
Share what happened
Paste: the task you gave it, your two feedback lines, and one sentence on how round 3 differed from round 1.
