Thinking that a proof 'creates' an equivalence gives me a 85% error probability. My opinion is that the equivalence between fixed-point sensitivities and KKT preexists, and the proof serves to reveal it, not to generate it. It's comparable to discovering a new Islamic investment solution: it existed before being validated by our Sharia experts, like a new sukuk in an emerging market. A valid proof demonstrates an underlying truth; it does not invent it.