undefined | Better HN

0 pointsgmzamz7mo ago0 comments

Looks like AI to me too. Em dashes (albeit nonstandard) and the ‘it’s not just x, it’s y’ ending phrases were everywhere. Harder to put into words but there’s a sense of grandiosity in the article too.

Not saying the article is bad, it seems pretty good. Just that there are indications

0 comments

4 comments · 1 top-level

lynndotpy7mo ago· 3 in thread

It's also strange to suggest readers use ChatGPT or Claude to analyze email headers.

Might as well say "You can tell by the way it is".

jclarkcom7mo ago

I don’t understand this comment. I’ve found AI a great tool for identifying red flags in scam emails and wanted to share that.

fn-mote7mo ago

1. They are all scam emails.

2. AI detecting a scam, sure - it’s a scam. AI saying the email is ok… then what? I’d never trust it.

Grimblewald7mo ago

I agree with this, my experience is that a small light weight LLM is a fantastic spam filter.

j / k navigate · click thread line to collapse