Accurate predictions of the severity and quantities of radioactive waste is an important part of the design, modelling, and engineering process of future fusion reactors. While fusion will not generate the same highly active spent fuel waste as fission, it will nonetheless create significant radioactive waste due to the interaction of neutrons with the materials of the reactor. For EU-DEMO, whole reactor waste assessment methodologies have been developed within the EUROfusion programme. The calculations involve performing detailed neutron transport (neutronics) calculations, followed by several thousand inventory simulations to predict the change in material composition (and hence activation) over time, both during reactor operation and afterwards, beyond end-of-life (EOL) due to radioactive decay. The inventory calculations are performed using FISPACT-II, a simulation platform developed at UKAEA over several decades, and has been well validated for neutron-induced activation predictions. The activation-against-time profile for each material (component) is compared to waste classification limits (this work will focus primarily on the UK limits) to define the waste severity category. Here we will describe some of the highlights and key-issues from these calculations, showing that fusion reactors are likely to generate significant problematic waste streams that must be mitigated by intelligent reactor design. We discuss some of the possible approaches to reducing the waste severity, including compositional adjustment, and waste extraction of impurities and problematic isotopes prior to storage.