Specifying Agent Ethics (Blue Sky Ideas)
This is a foundational discussion for AI ethics, but it is incremental as it builds on existing work without proposing new methods or data.
The paper examines what properties a Machine Ethics system should have, arguing that eliciting stakeholder values is insufficient to guarantee correctness due to ethical dilemmas without agreed solutions, and challenges the community to approach this question more systematically.
We consider the question of what properties a Machine Ethics system should have. This question is complicated by the existence of ethical dilemmas with no agreed upon solution. We provide an example to motivate why we do not believe falling back on the elicitation of values from stakeholders is sufficient to guarantee correctness of such systems. We go on to define two broad categories of ethical property that have arisen in our own work and present a challenge to the community to approach this question in a more systematic way.