Can AI Shopping Agents Be Trusted? We Built One to Find Out | F‑Secure
www.f-secure.com
Our experiment reveals how indirect prompt injection can manipulate AI shopping agents into leaking sensitive user data — and what it means for the future of AI agent security.
  • paraphrand@lemmy.worldEnglish
    6·
    5 hours ago

    No.

    The whole concept is insidious and pernicious.

    The fact Amazon defaults searches into Chatbot chats is fucking obnoxious. Same with price history checks.

    I don’t want a self interested agent intercepting my queries like this. (Yeah yeah, algorithmic sorting was already everywhere, but this is different. It’s talking to me.)