I consider the possibility that ultra-high-energy cosmic rays are accelerated in gamma-ray bursts located in the Galactic corona, thus circumventing the problem raised by Greisen-Zatsepin-Kuz'min cut-off The acceleration of UHECRs could occur in the pulsars which, in the coronal GRB model, produce them: the same parameters that permit fitting of GRB observations in the model of Podsiadlowski, Rees & Ruderman lead to an estimate of the highest achievable energies corresponding to that of the Bird et al. event, and to very low luminosities in cosmic rays. I show that, if the observations of Milgrom & Usov are confirmed, the extragalactic GRB model for the acceleration of UHECRs is untenable, but the same constraint does not apply to the coronal model. Also, I show that the efficiency of particle acceleration needs to be much smaller (and less demanding) than in cosmological models; of GRBs. Uncertainties remain about the ensuing cosmic ray spectral distribution. I also briefly discuss observational strategies to distinguish between the two possibilities.