How AI agents work: an AI that can only write, and still gets things done
AI agent permissions: who decides what an agent may do
An AI agent can write a message and send it to your friends before you see it. So who decides if a tool may run: the model, the app or you?
-
1
Go back to round three, the moment the model wrote its request to the messages tool. Nothing has gone out yet.
-
2
The model writes whatever request fits best, and it can be wrong: a forecast misread, a message to the wrong person. Nothing makes sure it stops itself. So what can?
The rest of this episode is for subscribers
Subscribe to read every premium episode. You can cancel when you want.
Paid yearly: $99 instead of $119.88