Passivation oxide formation is the key for corrosion control of metal alloys. The kinetics of competing oxide formation and dissolution determines alloy corrosion behaviors in aqueous solution. Despite the important role of the multi-component oxide evolution, little has been known on the kinetics from the atomistic level. We have built a computational framework that enables simulations of competing kinetic processes in multi-component oxides from first principles. The effects of applied voltage, pH and temperature on oxide growth, dissolution and reprecipitation can all be captured in this model. Combining with our experimental measurements on Alloy 22 and a Ni80%-Cr20% model alloy, we identified three voltage regimes with distinct oxide thicknesses and compositions. The oxide energetics of various stoichiometries are calculated by the density functional theory (DFT). Then the obtained data are used to train a surrogate lattice Hamiltonian with the cluster expansion (CE) method. Finally, kinetic Monte Carlo (KMC) simulations are run with the cation hopping barriers calculated on-the-fly based on the local environments from the combination of the above Hamiltonian and the linear Brønsted−Evans−Polanyi (BEP) relation.