Conference Paper/Proceeding/Abstract 6 views
Values, Preferences, and the Root of Moral Conflict
Fourth International Workshop on Value Engineering in AI (VALE 2026)
Swansea University Authors:
Jay Paul Morgan , Adam Wyner
Abstract
In Value-based Reasoning frameworks, values are used to guide the behaviour of agents and provide them with the means for creating actions that fulfil those values. In this article, we provide a logical formulation for how these values may be used in conjunction with the creation of goals (being pro...
| Published in: | Fourth International Workshop on Value Engineering in AI (VALE 2026) |
|---|---|
| Published: |
|
| URI: | https://cronfa.swan.ac.uk/Record/cronfa72061 |
| Abstract: |
In Value-based Reasoning frameworks, values are used to guide the behaviour of agents and provide them with the means for creating actions that fulfil those values. In this article, we provide a logical formulation for how these values may be used in conjunction with the creation of goals (being propositions about the environment the agent wishes to achieve). Through this formulation, we make use of preferences which allow for a ranking of actions with respect to the agent's goal, and give a discussion on the nature of Moral Conflict which arises when no single action respects all the agent's values. Within this discussion, we highlight the importance of Moral Conflict, and methods for resolving these when they occur. |
|---|---|
| Keywords: |
AI agents; Values; Reasoning; Value-based Decision Making |
| College: |
Faculty of Science and Engineering |

