Just checked with the team here, and it’s not us creating these accounts and data. Not sure who’s responsible or whether it’s malicious or not, but it’s not particularly helpful as has already been mentioned 
The main takeaway I guess will be that we need to accommodate for such scripts being run in the upcoming testnets, but that shouldn’t really hurt the rate of development I expect - it’ll probably just mean some more limitations added.
As always, thanks for the heads-up on this guys!