It's definitive that you have to give up locality (Bohmian mechanics), counterfactual definiteness (Copenhagen and others), or statistical independence (Superdeterminism). It's not definitive at all about local hidden variables.
Bell's theorem assumes statistical independence in its proof that local hidden variables can't reproduce QM. Superdeterminism violates statistical independence, therefore Bell's theorem does not rule out a superdeterministic local hidden variables theory, like this one:
