Structural design is often governed by climatic actions, such as snow, wind, thermal and atmospheric icing loads, that will occur during the design service life. Since in structural standards climatic actions are usually derived from historical data series assuming stationary climate, alterations induced by climate change should be specifically evaluated, also to assess their influence on structural reliability. In the paper, a probabilistic methodology for the assessment of climate change impact on long-term structural reliability is presented, based on the analysis of observed data series and climate projections, provided by high resolution climate models. Factor of change uncertainty maps for climate extremes are derived starting from the analysis of weather series generated by an ad hoc weather generator, which considers homogenous populations of data suitably derived from climate model output. The long-term structural reliability is then assessed for reference structures at a given site considering the non-stationary nature of climatic actions by means of the pdfs of changes in extreme value parameters. Specifically, variations of the failure probability with time due to climate change are evaluated by moving time windows of forty years considering changes in mean load intensity and standard deviation of yearly maxima of the investigated climatic action. The results show the capability of the method to assess the impact of climate change on structural safety, highlighting the necessity of adaptation measure to maintain the required target reliability of the structure during its life.