Whatup funnynoodles, there definitely are reasons this COULD end up bad, which are pretty much just that YC gets stretched too thin. But, honestly, it could also go super well for them. I mean, accelerators are entirely based on the idea of funding a lot of ideas to cherry pick the few good ones. Well, YC just accidentally created a batch of 15,000 ideas they are associated with, which could potentially provide some crazy successful startups. The downside is they lose money, and maybe some of the cohort are unhappy, though as they've told the later group they wouldn't have the same advisory resources, the program degradation feels fairly minimal.
Experiment - it could go bad, but it could also go good, we'll see.