For large-scale distribution and use of energy carriers classified as hazardous material in many countries as a method to assist land use planning, to grant licenses, to design a safe installation and to operate it safely some form of risk analysis and assessment is applied. Despite many years of experience the methods have still their weaknesses even the most elaborated ones as e.g. shown by the large spread in results when different teams perform an analysis on a same plant as was done in EU projects. Because a fuel as hydrogen with its different properties will come new in the daily use of many people incidents may happen and risks will be discussed. HySafe and other groups take good preparatory action in this respect and work in the right direction as appears from various documents produced. However, already a superficial examination of the results so far tells that further cooperative work is indispensable. To avoid criticism, skepticism and frustration not only the positive findings should be described and general features of the methods but the community has also to give strong guidance with regard to the uncertainties. Scenario development appears to be very dependent on insight and experience of an individual analyst, leak and ignition probability may vary over a wide range of values, Computational Fluid Dynamics, or CFD models may lead to very different result. The Standard Benchmark Exercise Problems, SBEPs, are a good start but shall produce guidelines or recommendations for CFD use or even perhaps certification of models. Where feasible narrowing of possible details of scenarios to the more probable ones taking into account historical incident data and schematizing in bowties, more explicit use of confidence intervals on e.g. failure rates and ignition probability estimates will help. Further knowledge gaps should be defined.