Apple Music uses a combination of machine learning and human curation to build personalized playlists and recommendations for each listener. Features like New Music Mix, Favorites Mix, and Get Up! Mix refresh automatically each week and are unique to each user, based on their listening history, songs they have "Loved," and tracks they have added to their library. These features have been part of Apple Music since its launch in 2015 and have grown more sophisticated over time.

Apple Music's recommendation system tracks engagement signals including plays, skips, library additions, and explicit "Love" ratings to build a picture of each user's taste. The platform uses machine learning methods to find patterns across millions of users — if two people consistently listen to the same artists, the system uses that to surface new music one user might not have found yet. Apple employs over 1,000 human music editors worldwide who curate editorial playlists and whose decisions are informed by algorithmic data. Apple has not publicly released technical documentation on its recommendation architecture, making the precise methods reported rather than officially detailed.