AI now writes phishing without the spelling and grammar errors that once gave it away. The modern defense is questioning whether a request makes sense and verifying it, rather than scanning for mistakes that no longer appear.
“Typos used to catch phishing. Now you have to question intent.”