Unproven computational hardness assumption that 3-SAT isn’t solvable in subexponential time in the worst case