The linguistic dead zone of value-aligned agency, natural and artificial

LaCroix, Travis

Abstract

The value alignment problem for artificial intelligence (AI) asks how we can ensure that the “values”—i.e., objective functions—of artificial systems are aligned with the values of humanity. In this paper, I argue that linguistic communication is a necessary condition for robust value alignment. I discuss the consequences that the truth of this claim would have for research programmes that attempt to ensure value alignment for AI systems—or, more loftily, those programmes that seek to design robustly beneficial or ethical artificial agents.

Citation

LaCroix, T. (online). The linguistic dead zone of value-aligned agency, natural and artificial. Philosophical Studies, https://doi.org/10.1007/s11098-024-02257-w

• Journal Article Type: Article
• Acceptance Date: Nov 8, 2024
• Online Publication Date: Dec 4, 2024
• Deposit Date: Dec 5, 2024
• Publicly Available Date: Dec 4, 2024
• Journal: Philosophical Studies
• Print ISSN: 0031-8116
• Electronic ISSN: 1573-0883
• Publisher: Springer
• Peer Reviewed: Peer Reviewed
• DOI: https://doi.org/10.1007/s11098-024-02257-w
• Keywords: Artificial intelligence, AI, The value alignment problem, Principal-agent problems, Machine learning, Objective functions, Normative theory, Language, Linguistic communication, Communication systems, Information transfer, Coordination, Values, Preferences
• Public URL: https://durham-repository.worktribe.com/output/3201272