I don't have technical knowledge on neuromorphic computing (NC) but occasionally listen to a podcast (EETIMES current) dedicated to frontier research on the field. My impression is large scale application of NC is still working on the proof of concept, for example it is commonly quoted that NC has energy or performance advantage on some task based on some small scale calculation, but they generally lacks of rigorous theoretical framework to prove which types of works they are good at. They have good progress in scaling up number of neurons in NC, but the software ecosystem is still very immature, there's missing gap between low level and high level language so it is difficult to bring researchers worldwide to work on the ecosystem. But certainly not a dead end since not much resources (e.g. compare to quantum computing) have been given to it anyway, and it seems like an exciting and rapidly expanding field from academic perspective.