EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Using the Veil of Ignorance to align AI systems with principles of justice

Laura Weidinger, Kevin R McKee iD, Richard Everett, Saffron Huang, Tina O Zhu, Martin J Chadwick, Christopher Summerfield iD, Iason Gabriel

Proceedings of the National Academy of Sciences · 2023

DOI: 10.1073/pnas.2213709120 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the National Academy of Sciences, 2023. Verfasst von Laura Weidinger und 7 weiteren Autor:innen. Beteiligte Einrichtung: DeepMind, London N1C 4DN, United Kingdom. Hauptthemen laut MEDLINE: Artificial Intelligence, Societies. Die Arbeit wurde bisher 44-mal zitiert (Crossref) und nennt 75 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The philosopher John Rawls proposed the Veil of Ignorance (VoI) as a thought experiment to identify fair principles for governing a society. Here, we apply the VoI to an important governance domain: artificial intelligence (AI). In five incentive-compatible studies ( N = 2, 508), including two preregistered protocols, participants choose principles to govern an Artificial Intelligence (AI) assistant from behind the veil: that is, without knowledge of their own relative position in the group. Compared to participants who have this information, we find a consistent preference for a principle that instructs the AI assistant to prioritize the worst-off. Neither risk attitudes nor political preferences adequately explain these choices. Instead, they appear to be driven by elevated concerns about fairness: Without prompting, participants who reason behind the VoI more frequently explain their choice in terms of fairness, compared to those in the Control condition. Moreover, we find initial support for the ability of the VoI to elicit more robust preferences: In the studies presented here, the VoI increases the likelihood of participants continuing to endorse their initial choice in a subsequent round where they know how they will be affected by the AI intervention and have a self-interested motivation to change their mind. These results emerge in both a descriptive and an immersive game. Our findings suggest that the VoI may be a suitable mechanism for selecting distributive principles to govern AI.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Artificial IntelligenceSocietiesHumansSocial Justice

Stichwörter der Autor:innen

artificial intelligencedecision-makingethicsfairnesspolicymaking

Autor:innen und Institutionen

  • DeepMind, London N1C 4DN, United Kingdom.
  • Department of Psychology, University of Oxford, Oxford OX2 6GG, United Kingdom.

ORCID-Profile: Kevin R McKee, Christopher Summerfield

Zitieren

APA

Weidinger, L., McKee, K. R., Everett, R., Huang, S., Zhu, T. O., Chadwick, M. J., Summerfield, C., & Gabriel, I. (2023). Using the Veil of Ignorance to align AI systems with principles of justice. Proceedings of the National Academy of Sciences. https://doi.org/10.1073/pnas.2213709120

Vancouver

Weidinger L, McKee KR, Everett R, Huang S, Zhu TO, Chadwick MJ, et al. Using the Veil of Ignorance to align AI systems with principles of justice. Proceedings of the National Academy of Sciences. 2023. doi:10.1073/pnas.2213709120

BibTeX
@article{Weidinger2023,
  author = {Weidinger, Laura and McKee, Kevin R and Everett, Richard and Huang, Saffron and Zhu, Tina O and Chadwick, Martin J and Summerfield, Christopher and Gabriel, Iason},
  title = {Using the Veil of Ignorance to align AI systems with principles of justice},
  journal = {Proceedings of the National Academy of Sciences},
  year = {2023},
  doi = {10.1073/pnas.2213709120}
}
Kostenlos starten →