It cannot be that profitable to have just a bunch of random data on their servers. I have so much junk and random bullshit on my drives, it would take a week of labor just to clean my shit well enough to use it for AI training and as soon as I got any notification about cloud space being full i'd turn syncing off - i sure as hell wouldn't fork over any money for a subscription. This is such a big bridge to burn, and the server overhead must be massive.... I just don't understand how they could possibly think this is a good business decision.
Idk, maybe i'm just too deep into privacy/FOSS/selfhosting headspace to see things clearly from the normal-consumer standpoint but I just do not understand this. I really wish someone would leek an internal conversation at one of these companies that explains the big-picture strategy with this move.