Towards Responsibly Non-Compliant Machines
This work addresses the need for ethical AI that can appropriately disobey harmful or unethical commands, but it is a conceptual sketch rather than a concrete solution.
The paper explores the concept of autonomous agents that can responsibly refuse user requests, outlining key issues such as justification for refusal, override mechanisms, and tracking of security risks and liability.
We consider the problem of engineering autonomous intelligent agents that are capable to responsibly not comply with user requests. We argue that machine non-compliance comes in many different forms, and sketch the issues we should pursue on the road of accomplishing responsibly non-compliant intelligent machines. We anchor responsible non-compliance in justifications for task refusal, pathways to override the non-compliance, as well as careful tracking of security risks and liability transfers.