It’s more of a:

“If person A acts badly, Person B to ∞ act badly to person A”.

But I can see the programming failure of the suggested rule as written, as it can be implied that only person B should act upon person A, rather than everyone.