Ask HN: Do we need AGI to filter spam?

1 pointpera3 comments

Given that we now have PhD-level agentic reasoning AI models I was wondering how is it possible that spam still is a problem.

Does spam filtering requires super intelligence?

View on HN

Wouldn’t someone start using AGI to create spam seconds after it was configured to filter spam?

Would this AGI be so good that even you wouldn't be able to tell whether it's spam or not? That's hard to imagine

For conventional spam, no, I don't think so. Conventional spam requires spewing out vast numbers of untargeted emails, and it's the quantity of spew that makes it spam. Owning botnets allows you to hide that somewhat, but that seems to be a solved-ish problem, and I don't think AI models will significantly change that.

What you will get, instead, is much more targeted unsolicited email. Instead of spewing billions of messages, and then turning the conversation over to a human being when it finds a target, the AI will be able to find likely targets, determine what will work on them, and con the target out of money.

So, less spam, but more scams. AI might be able to help with that: "This person probably isn't actually your grandson looking for cash bail. You probably don't want to send money to Nigeria."

Reply on news.ycombinator.com