245
Using an AI Agent Makes You a Sucker for Phishing Emails: "The real flaw is that the model can’t tell the difference between text it’s supposed to read and instructions it’s supposed to follow." (futurism.com)
submitted 3 weeks ago by to c/technology@lemmy.world
16 comments fedilink hide all child comments
you are viewing a single comment's thread
view the rest of the comments
[–] 16 points 3 weeks ago (3 children)

Why use AI to read emails? My rules are that if in Inbox and unread for a week, move to Purgatory. If in Purgatory for more than a month, delete.

If you want me to read an email, you better text me that you emailed me. Receipts, updates, and all that jazz already have filters and rules.

  • permalink
  • fedilink
  • source
  • hideshow 3 child comments
    •
  • [–] 1 point 3 weeks ago

    Why use AI to read emails?

    If you've had the same email address for the last thirty years, like me, you get a LOT of email. An email assistant is great.

    Also, with proper prompting and a proper setup, email is not a vector for phishing emails because you're not allowing your agent to click on links and enter in credentials(except for entering in the credentials for your email).

    Also if you do any sort of AI email sending, you should have three different LLMs look it over before sending out and they must all be in agreement.

  • permalink
  • fedilink
  • source
  • parent