but if you introduce a derivative which is covariant under diffeomorphisms and u(1) transformations
D = d + \omega + A
you see that gravitation (connection \omega) enters on a somewhat same footing as the electromagnetic potential (A). there are nuances ofc, but the view (force/curvature) depends from where you're looking.
also in real life particles don't real exist... they're excitations in a field