Technology Category
Humans and AI: Learning Human Values and PreferencesPhilosophy and Ethics of AI: Morality & Value-based AIMultiagent Systems: Agreement, Argumentation & Negotiation
Rethinking How AI Embeds and Adapts to Human Values: Challenges and Opportunities
🤖 AI Summary
This paper addresses a core challenge in AI value alignment: the dynamic, pluralistic, and often conflicting nature of human values, coupled with the limitations of prevailing static alignment paradigms in long-term adaptability and inter-agent negotiation. We propose a *Dynamic Value Alignment* framework that models value embedding as an evolvable process, integrating multi-agent system modeling, dynamic preference learning, cross-population value inference, and value-sensitive design. Our key contributions are threefold: (1) the first systematic incorporation of multi-agent negotiation mechanisms to resolve value tensions across individuals and groups; (2) explicit emphasis on long-horizon reasoning and adaptive value evolution; and (3) deliberate interdisciplinary theoretical integration to encompass the full spectrum of human values. The work rigorously characterizes fundamental challenges and outlines a scalable, inclusive, and human-centered research agenda—advancing both the theoretical foundations and practical deployment of human-aligned AI.
Technology Category
Application Category
📝 Abstract
The concepts of ``human-centered AI'' and ``value-based decision'' have gained significant attention in both research and industry. However, many critical aspects remain underexplored and require further investigation. In particular, there is a need to understand how systems incorporate human values, how humans can identify these values within systems, and how to minimize the risks of harm or unintended consequences. In this paper, we highlight the need to rethink how we frame value alignment and assert that value alignment should move beyond static and singular conceptions of values. We argue that AI systems should implement long-term reasoning and remain adaptable to evolving values. Furthermore, value alignment requires more theories to address the full spectrum of human values. Since values often vary among individuals or groups, multi-agent systems provide the right framework for navigating pluralism, conflict, and inter-agent reasoning about values. We identify the challenges associated with value alignment and indicate directions for advancing value alignment research. In addition, we broadly discuss diverse perspectives of value alignment, from design methodologies to practical applications.
Problem
Research questions and friction points this paper is trying to address.
Rethinking AI embedding of human values in systems
Addressing risks of harm from value misalignment
Adapting AI to evolving and pluralistic human values
Innovation
Methods, ideas, or system contributions that make the work stand out.
Implementing long-term reasoning for value adaptation
Using multi-agent systems for pluralistic value navigation
Developing theories for full spectrum human values
🔎 Similar Papers
No similar papers found.
