This paper examines the potentially important role of generative AI (GenAI) technology in the domain of knowledge discoveries in biomedicine. The purpose of the paper is to experiment, using ChatGPT, if we could enhance and accelerate our traditional way of discovering knowledge in biomedicine, by (a) asking questions through prompting and obtaining answers which deliver new knowledge related to the question and (b) managing prompting with the expert knowledge of a particular problem domain. The results of the experiments are interesting. Text generation with GenAI technologies is here to stay and can contribute toward faster and more efficient discoveries of facts, research results, and advances in biomedical science. However, it is debatable if we can rely on the results of this type of text generation, because of the level of GenAI precision, the nature of large language model tuning, and the impact of expert knowledge on prompt engineering.