Preimplantation genetic screening (PGS) has been proposed as a method to improve the success rates of assisted conception in certain indications. Most commonly the technique involves analysing part of the karyotype of one or two biopsied blastomeres by fluorescence in situ hybridisation and thus detecting aneuploid embryos which are then not transferred into the uterus. Proposed indications include advanced maternal age, repeated IVF failure and recurrent miscarriage. While there are a considerable number of reports of the use of this technique, there are only two randomised controlled trials reported at present, both for the indication of advanced maternal age. Neither show an increase in live birth rate, and indeed the more recent (using a relatively low age cut-off of 35 years) suggests that live birth rate is reduced by the use of PGS. Methodological aspects of both studies are discussed. It remains possible that PGS may be of benefit under certain circumstances. However at present patients should be informed that there is no robust evidence that PGS for advanced maternal age improves live birth rate per cycle started, and PGS should preferably be offered within the context of robustly designed randomised trials performed in suitably experienced centres.