The article discusses the so-called 'kT problem' with its formulation, content, and consequences. The usual formulation of the problem points out the paradox of biological effects of weak low-frequency magnetic fields. At the same time, the formulation is based on several implicit assumptions. Analysis of these assumptions shows that they are not always justified. In particular, molecular targets of magnetic fields in biological tissues may operate under physical conditions that do not correspond to the aforementioned assumptions. Consequently, as it is, the kT problem may not be an argument against the existence of non thermal magnetobiological effects. Specific examples are discussed: magnetic nanoparticles found in many organisms, long-lived rotational states of some molecules within protein structures, spin magnetic moments in radical pairs, and magnetic moments of protons in liquid water.