I expect that with AI, we'll be less able to rely on the heuristic of bad grammar to easily detect phishing. That one flaw gave the phishers away so often, and made it so obvious ...
I suspect it's a compromise. Entertaining a target with high literacy skills is more work. If you have AI that is excellent at communication, there goes one reason to aim low.
I see this a lot, but it seems hard to verify. Has a scammer actually ever come out and say they deliberately use poor grammar? Would be interesting to compare scam emails/messages from scammers based in the US/UK/Australia with those in India or Nigeria, and see if the pattern holds up for both.
Easy. As you say all you need to do is to compare and as I occasionally have a glimpse of the spam for some non-English language with a non-Latin script - it's the same.
Absolutely. I was talking to a teacher and they said the translations were easy to spot because they were so good, far beyond what the kid was normally doing. I then showed them the same kind of prompt plus something like "write this as an X year old French student with only moderate grasp of ..." and it was far more plausible.
I noticed how well it understands general parlance after it created marketing style copy for me and I told it to sound "less wanky" and it made it much more to the point.
