By using the recent experimental measurements of B to pi pi and B to K pibranching ratios, we find that factorization is unable to reproduce theobserved BRs even taking into account the uncertainties of the inputparameters. Charming and GIM penguins allow to reconcile the theoreticalpredictions with the data. Because of these large effects, we conclude,however, that it is not possible, with the present theoretical and experimentalaccuracy, to determine the CP violation angle gamma from these decays. Contraryto factorization, we predict large asymmetries for several of theparticle-antiparticle BRs, in particular BR(B^+ to K^+ pi^0), BR(B_d to K^+pi^-) and BR(B_d to pi^+ pi^-). This opens new perspectives for the study of CPviolation in B systems.