alt Hacker NewsIt's a very comparable game of cat and mouse to spam email filtering. People also tried to claim that spam was over because for a time companies like Google cared enough to invest a lot in preventing as much as possible from getting through. If you've noticed in recent years the motivation to keep up that level of filtering has greatly diminished.
Whether model poisoning becomes a bigger issue depends on the incentives for companies to keep fighting it. For now in comparison to attackers the incentives and resources needed to defend against model poisoning are huge so it's just temporary setbacks. Will that unevenness in their favor always be the case?
Replies
>It's a very comparable game of cat and mouse to spam email filtering. People also tried to claim that spam was over because for a time companies like Google cared enough to invest a lot in preventing as much as possible from getting through. If you've noticed in recent years the motivation to keep up that level of filtering has greatly diminished.
https://en.wikipedia.org/wiki/Lotka%E2%80%93Volterra_equatio...
I feel like spam filtering has moved from statistical methods to pay-to-play: "These 10 large senders have a reasonable opt-out policy (on paper, we'll check any day now), so why would we filter anything they drop at our 25?"