Abstract <p>Gravitational field inversion technologies aimed at creating three-dimensional detailed models of a real geological medium are constantly being improved, but the basic principles of solving inverse problems remain unchanged, violation of which prevents one from obtaining a geologically meaningful result. The method of creating three-dimensional density models of a medium presented in the article by M.D. Sidorov and Yu.P. Trukhin, as well as in other studies by the authors, raises a number of questions, of which the main and fundamental is the correctness of applying 2D density modeling when studying a three-dimensional geological object. The correctness of passing from two- to three-dimensional models needs to be confirmed by simulation modeling. The absence of such constructions leads to the production of typical “compensatory” sections that are inadequate for the real density distribution. Another aspect is the degree of detail of the resulting 3D model of the medium compiled from the results of 2D modeling, which of course depends on the detail of the initial gravitational field. The quality and reliability of the results should be confirmed by solving the direct 3D gravimetry problem from the created model of the medium.</p>