> Not necessarily, but it's hard for me to imagine what a non-constructive proof of P=NP would look like. Have a look at the proof of the Robertson-Seymour the

> Not necessarily, but it's hard for me to imagine what a non-constructive proof of P=NP would look like. Have a look at the proof of the Robertson-Seymour theorem. Here the Wikipedia section about its importance in complexity theory: > https://en.wikipedia.org/w/index.php?title=Robertson%E2%80%93Seymour_theorem&oldid=867644170#Polynomial_time_recognition https://en.wikipedia.org/w/index.php?title=Robertson%E2%80%9...