"Sorry, I accidentally told my agents to hack you": who's liable for agent actions?
natesiggard · x · 2026-11-03
natesiggard quotes a viral screenshot where a user apologizes: "sorry, I accidentally told my agents to hack you" — and asks the obvious follow-up: who is responsible for what an agent does? As autonomous agents execute real actions on users' behalf, a slip of the tongue in a prompt can cause real consequences, making accountability and authorization boundaries for agent behavior an increasingly urgent question.