Well, it's a minor point, but I think it's fair to say that getting to pick the arbitrary line on any behavior in a social context comes with some degree of power and, to borrow from Spider-Man, responsibility. Hypothetically speaking, if you set the line just above the creator's power, that would be hard to justify as not some kind of advantage, albeit only as large as dodging the bot would be. So, probably not much.
Have you considered adjusting the bot with only one change - no flag, just the reports/updates? It's too bad there's no PM feature, because hypothetically, you could even PM a warning before a public name-and-shame.
I'm not sure these are good suggestions, just off the top of my head.