I just realized that the SAT problem in complexity theory can be reformulated in my framework as the problem of finding a flat informational network — one with globally consistent holonomies. Since SAT is NP-complete, this implies that minimizing the physical action (defined as informational curvature) is generally computationally intractable. Proving P ≠ NP would then be equivalent to showing that a non-zero curvature gap always exists, which is precisely the statement of the Yang–Mills mass gap in physics. 🤯
