An intercomparison exercise is described that examines Monte Carlo modelling of anthropomorphic voxel phantoms in an idealized ground-contamination photon exposure scenario. Thirteen participants calculated and submitted organ and effective dose rates for comparison against a set of verified reference solutions. The effective dose rates are shown to agree with the reference value to within reasonable statistical uncertainties in five of the cases, though in only one of those was similar agreement also demonstrated in the evaluation of all requested organ dose rates. Orders-of-magnitude differences in doses are seen for some of the other participants, both internally within their own dataset and also relative to the reference solutions. Following limited feedback and suggestions from the organizer, up to two sets of revised solutions were resubmitted by some of the participants; these generally exhibited improved agreement, though not always. The overall observations and conclusions from this intercomparison exercise are summarized and discussed.