Positive AI: Key Challenges in Designing Artificial Intelligence for Wellbeing

Willem van der Maden; Derek Lomas; Malak Sadek; Paul Hekkert

doi:10.48550/arXiv.2304.12241

Positive AI: Key Challenges in Designing Artificial Intelligence for Wellbeing

Willem van der Maden
, Derek Lomas
, Malak Sadek
, Paul Hekkert

Research output: Other contribution › Research

Abstract

Artificial Intelligence (AI) is a double-edged sword: on one hand, AI promises to provide great advances that could benefit humanity, but on the other hand, AI poses substantial (even existential) risks. With advancements happening daily, many people are increasingly worried about AI’s impact on their lives. To ensure AI progresses beneficially, some researchers have proposed “wellbeing” as
a key objective to govern AI. This article addresses key challenges in designing AI for wellbeing. We group these challenges into issues of modeling wellbeing in context, assessing wellbeing in context, designing interventions to improve wellbeing, and maintaining AI alignment with wellbeing over time. The identification of these challenges provides a scope for efforts to help ensure that AI
developments are aligned with human wellbeing.

Original language	English
Publication date	2023
Publisher	arXiv
Number of pages	1
DOIs	https://doi.org/10.48550/arXiv.2304.12241
Publication status	Published - 2023
Externally published	Yes

Keywords

Artificial Intelligence
Wellbeing
Value-alignment
Positive Design
Positive computing
Cybernetics

Access to Document

10.48550/arXiv.2304.12241Licence: Unspecified

Cite this

@misc{3611398d51044e678af37e3d6fa842b7,

title = "Positive AI: Key Challenges in Designing Artificial Intelligence for Wellbeing",

abstract = "Artificial Intelligence (AI) is a double-edged sword: on one hand, AI promises to provide great advances that could benefit humanity, but on the other hand, AI poses substantial (even existential) risks. With advancements happening daily, many people are increasingly worried about AI{\textquoteright}s impact on their lives. To ensure AI progresses beneficially, some researchers have proposed “wellbeing” as a key objective to govern AI. This article addresses key challenges in designing AI for wellbeing. We group these challenges into issues of modeling wellbeing in context, assessing wellbeing in context, designing interventions to improve wellbeing, and maintaining AI alignment with wellbeing over time. The identification of these challenges provides a scope for efforts to help ensure that AI developments are aligned with human wellbeing.",

keywords = "Artificial Intelligence, Wellbeing, Value-alignment, Positive Design, Positive computing, Cybernetics",

author = "\{van der Maden\}, Willem and Derek Lomas and Malak Sadek and Paul Hekkert",

year = "2023",

doi = "10.48550/arXiv.2304.12241",

language = "English",

publisher = "arXiv",

type = "Other",

}

TY - GEN

T1 - Positive AI: Key Challenges in Designing Artificial Intelligence for Wellbeing

AU - van der Maden, Willem

AU - Lomas, Derek

AU - Sadek, Malak

AU - Hekkert, Paul

PY - 2023

Y1 - 2023

N2 - Artificial Intelligence (AI) is a double-edged sword: on one hand, AI promises to provide great advances that could benefit humanity, but on the other hand, AI poses substantial (even existential) risks. With advancements happening daily, many people are increasingly worried about AI’s impact on their lives. To ensure AI progresses beneficially, some researchers have proposed “wellbeing” as a key objective to govern AI. This article addresses key challenges in designing AI for wellbeing. We group these challenges into issues of modeling wellbeing in context, assessing wellbeing in context, designing interventions to improve wellbeing, and maintaining AI alignment with wellbeing over time. The identification of these challenges provides a scope for efforts to help ensure that AI developments are aligned with human wellbeing.

AB - Artificial Intelligence (AI) is a double-edged sword: on one hand, AI promises to provide great advances that could benefit humanity, but on the other hand, AI poses substantial (even existential) risks. With advancements happening daily, many people are increasingly worried about AI’s impact on their lives. To ensure AI progresses beneficially, some researchers have proposed “wellbeing” as a key objective to govern AI. This article addresses key challenges in designing AI for wellbeing. We group these challenges into issues of modeling wellbeing in context, assessing wellbeing in context, designing interventions to improve wellbeing, and maintaining AI alignment with wellbeing over time. The identification of these challenges provides a scope for efforts to help ensure that AI developments are aligned with human wellbeing.

KW - Artificial Intelligence

KW - Wellbeing

KW - Value-alignment

KW - Positive Design

KW - Positive computing

KW - Cybernetics

UR - http://www.scopus.com/inward/record.url?eid=2-s2.0-85188023181&partnerID=MN8TOARS

U2 - 10.48550/arXiv.2304.12241

DO - 10.48550/arXiv.2304.12241

M3 - Other contribution

PB - arXiv

ER -

Positive AI: Key Challenges in Designing Artificial Intelligence for Wellbeing

Abstract

Keywords

Access to Document

Fingerprint

Cite this