... hurt: a mass email to 50k users, or a migration you can only roll back after downtime. "Reversible on paper, ruinous in practice." His version is that the gate should key on reach, not on reversibility. Not "is this safe" but "how far does this reach if I'm wrong." And the part that stuck with me: the agent is the worst possible judge of its own blast radius, because it does not know there are 50k people on that ... ... agent does reaches past me. Genuinely low stakes, no gate needed. B. It could reach customers, I know it, and I have not built the gate yet. C. It stops and waits for someone to come back and look. Safe

... transfer for me to approve than execute one directly, not distrust of the model, just distrust of how hard undoing a mistake really is once money has moved. Comment from g023 - Developer(@g023): These days, I feel pretty safe with the right agents working the show, but its good to pay attention, and make sure any applications that you use for finances or server information are 2 factor tied to your phone or something (and logged out). Just ... ... agent economies where human oversight becomes even more critical. That could unlock insane productivity or create entirely new classes of misalignment risk. Where do you personally draw the line between helpful proactive assistant and too autonomous to feel safe