Landscape of Machine Implemented Ethics

Vivek Nallur
DOI: https://doi.org/10.1007/s11948-020-00236-y
2020-09-01
Abstract:This paper surveys the state-of-the-art in machine ethics, that is, considerations of how to implement ethical behaviour in robots, unmanned autonomous vehicles, or software systems. The emphasis is on covering the breadth of ethical theories being considered by implementors, as well as the implementation techniques being used. There is no consensus on which ethical theory is best suited for any particular domain, nor is there any agreement on which technique is best placed to implement a particular theory. Another unresolved problem in these implementations of ethical theories is how to objectively validate the implementations. The paper discusses the dilemmas being used as validating 'whetstones' and whether any alternative validation mechanism exists. Finally, it speculates that an intermediate step of creating domain-specific ethics might be a possible stepping stone towards creating machines that exhibit ethical behaviour.
Artificial Intelligence
What problem does this paper attempt to address?
The problem that this paper attempts to solve is the implementation difficulty in machine ethics, that is, how to achieve ethical behavior in robots, self - driving vehicles or software systems. Specifically, the paper focuses on the following aspects: 1. **Selection and implementation of ethical theories**: Currently, there is no consensus on which ethical theory is the most suitable for a specific application area, nor is it clear which technology is the most suitable for implementing a certain ethical theory. Therefore, the paper explores different ethical theories and their implementation technologies, aiming to find more appropriate solutions. 2. **Validation of the effectiveness of implementation technologies**: How to objectively verify the implementation effects of these ethical theories is an unsolved problem. The paper discusses whether the ethical dilemmas (such as the trolley problem) used for verification are effective and whether there are other verification mechanisms. 3. **Implementation of cross - domain ethics**: As the application range of autonomous systems continues to expand, ethical requirements in different fields also vary. The paper proposes that creating ethical norms in specific fields may be an intermediate step in implementing machine ethics, which is helpful for finally developing machines that can exhibit ethical behavior. Through these discussions, the paper aims to promote research in the field of machine ethics, ensure that the behavior of future autonomous systems complies with social ethical standards, and thus enhance human trust and acceptance of these systems.