Myth vs Reality ledger

Somebody's favourite myth, tested every issue

One popular claim about building software with AI agents per issue, put up against the evidence. Most don't survive. The ones that do get a fair hearing.

What people say

Don't worry, the agent asks before it runs anything. A human approves every command.

Myth

The human approves the command; the damage is in what the command sets off. Sandboxes and effect checks keep you safe. The click keeps you busy.

Last updated . Reuse with credit under CC BY 4.0: “The Dabbawala Protocol, thedabbawalaprotocol.com”.