From a Google/Alphabet perspective it could also be valuable to distinguish between „original“ and „ai generated“ music for the purpose of a cleaner database to train their own music generation models?