Pluralistic Alignment Over Time
Preprint
Abstract
If an AI system makes decisions over time, how should we evaluate how aligned it is with a group of stakeholders (who may have conflicting values and preferences)? In this position paper, we advocate for consideration of temporal aspects including stakeholders' changing levels of satisfaction and their possibly temporally extended preferences. We suggest how a recent approach to evaluating fairness over time could be applied to a new form of pluralistic alignment: temporal pluralism, where the AI system reflects different stakeholders' values at different times.Keywords
All Related Versions
This publication has 15 references indexed in Scilit:
- Beyond Preferences in AI AlignmentPhilosophical Studies, 2024
- What Properties of Norms can we Implement in Robots?*2023
- Perpetual Voting: Fairness in Long-Term Decision MakingProceedings of the AAAI Conference on Artificial Intelligence, 2020
- Norms, Rewards, and the Intentional Stance2018
- Authoring and Verifying Human-Robot Interactions2018
- Specifying and computing preferred plansArtificial Intelligence, 2011
- A mixed Bentham–Rawls criterion for intergenerational equity: Theory and implicationsJournal of Environmental Economics and Management, 2009
- A heuristic search approach to planning with temporally extended preferencesArtificial Intelligence, 2009
- Planning with Preferences Using Logic ProgrammingLecture Notes in Computer Science, 2003
- Planning for temporally extended goalsAnnals of Mathematics and Artificial Intelligence, 1998