The paper suggests an application of the precautionary principle to the use of genetics in psychiatry focusing on scientific uncertainty. Different levels of uncertainty are taken into consideration--from the acknowledgement that the genetic paradigm is only one of the possible ways to explain psychiatric disorders, via the difficulties related to the diagnostic path and genetic methods, to the value of the results of studies carried out in this field. Considering those uncertainties, some measures for the use of genetics in psychiatry are suggested. Some of those measures are related to the conceptual limits of the genetic paradigm; others are related to present knowledge and should be re-evaluated.