The greatest challenge in the interpretation of galaxy clustering data from any surveys is galaxy bias. Using a simple Fisher matrix analysis, we show that the bispectrum provides an excellent determination of linear and non-linear bias parameters of intermediate and high-z galaxies, when all measurable triangle configurations down to mildly non-linear scales, where perturbation theory is still valid, are included. The bispectrum is also a powerful probe of primordial non-Gaussianity. The planned galaxy surveys at z 2 should yield constraints on non-Gaussian parameters, f loc.N L and f eq. N L , that are comparable to, or even better than, those from CMB experiments. We study how these constraints improve with volume, redshift range, as well as the number density of galaxies. Finally, we show that a halo occupation distribution may be used to improve these constraints further by lifting degeneracies between gravity, bias, and primordial non-Gaussianity.