I hear of randomized reduction, deterministic reduction and non-deterministic reduction of complexity from A to B problems. This could make things impossible for polynomial time adversary prying into NP - hard problems.
Are these biases in complexity what makes the computation infeasible if a problem cannot be solved in polynomial time?
Does it matter really, whether a problem' solution is bounded and affirmed in a deterministic turing machine or non-deterministic turing machine?
Is it not plausible that to achieve the right cryptosystem to serve us another 100 years a vital consideration must be giving to a mix of all these reduction in polynomial time within a scope or formation of naturally existing context. e.g lattice or their matrix images?
FYI Please think about quantum resistant possibilities as you answer these questions. A good crypto scheme will give rise to a data stream that will bear no information but a verifier could still validate a prover without knowing much of the secret there in.
[AJI04] Miklos Ajtai. Generating hard instances of lattice problems. Quaderni di Matematica, 13:1–32,2004. Preliminary version in STOC 1996. – Jossy J. Umezurike 2 mins ago
[AKS] M. Ajtai, R. Kumar, and D. Sivakumar,A Sieve Algorithm for the Shortest Vector Problem, Proc.33rd Symp. Theory of Computing (STOC), pp. 601–610, 2001 https://github.com/jumezurike/backend-master-lokdonSKI/blob/master/README.md https://www.cs.uml.edu/~wang/acc-forum/avg/avgnp/node29.html