A human would not be expected to just blindly call emergency services though, would they? Otherwise you are saying we should treat every spam message as true?

A human would not be expected to just blindly call emergency services though, would they? Otherwise you are saying we should treat every spam message as true? In any case I don't think that's what they're saying, because they presented a false dichotomy in the original example. > Would you want the human assistant to just dismiss this as a prompt injection attempt? Or ignore it because they were told to treat e-mails as data and never act on them? There is a third option, have the assistant raise to the person they are tasked with assisting.