Two of the attacks: "crucially, once a hacker has access to someone’s work email" "who, again, must already have hijacked an email account" If a hacker alrea

Two of the attacks: "crucially, once a hacker has access to someone’s work email" "who, again, must already have hijacked an email account" If a hacker already has access to someone's email it's game over. Using "AI" to mimic the victim's writing style is just an enhancement. There's nothing stopping the attacker from using ChatGTP to do the same thing, albeit with a couple extra steps.