Thanks to machine learning and, in particular, deep learning methods using artificial neural networks and the increasing availability of digital health data, the diagnosis and treatment of diseases is making great progress. People suffering from rare cancers can also benefit from such approaches that work with artificial intelligence (AI). Diagnosis and treatment of rare cancers are facing significant developments in view of the new AI methods, but also numerous technical and non-technical challenges. The use of AI applications for the detection and treatment of rare cancers is one aspect of the discussion on how to deal with increasingly personalised medicine. By taking into account the individual genetic, physical and morphological characteristics of patients, the chance of a precisely tailored treatment that is as effective as possible and has as few side effects as possible increases. However, such individualisation of therapeutic approaches - whether as a drug or medical device - is difficult to fit into the framework of existing approval regimes, which have so far been based as far as possible on extensive clinical trials with numerous test subjects, standardised products and standardised procedures. These procedures are more difficult to implement for rare cancers. Experimental treatments are permitted within the scope of therapeutic freedom and as therapeutic trials within narrow limits in order to treat individual patients with novel and less tested approaches. Nevertheless, a reliable framework is necessary to prove the effectiveness of personalised diagnosis and therapy. This also applies to AI-based diagnostic and therapeutic procedures. To a certain extent, the AI systems themselves become unique instruments, as they represent an individualised version of their artificial neural networks according to the data used in the specific application context. In principle, it is to be expected that the treatment of rare cancers will also benefit from advances in the availability and use of digital health data, adaptive AI applications and personalised diagnostics and therapy in general oncology.