The TopCat AI Signal Report #89

THE TOPCAT FRAMEWORK — BOUNDED AUTONOMY

So I’m adding a simple framework to our Signal vocabulary:

MISSION

What is the agent trying to accomplish?

ACCESS

What information can it see?

AUTHORITY

What can it change?

LIMITS

What can it never do?

ESCALATION

When must it ask a human?

AUDIT

Can we reconstruct what happened?

SHUTDOWN

Can we stop it immediately?

That’s the architecture of bounded autonomy.

And I think we’re going to hear that concept more and more.