Medical Artificial Intelligence and Human Values

Key PointsMedical Artificial Intelligence and Human ValuesAs large language models and other artificial intelligence models are used more in medicine, ethical dilemmas can arise depending on how the model was trained. A user must understand how human decisions and values can shape model outputs. Med...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:The New England journal of medicine 2024-05, Vol.390 (20), p.1895-1904
Hauptverfasser: Yu, Kun-Hsing, Healey, Elizabeth, Leong, Tze-Yun, Kohane, Isaac S., Manrai, Arjun K.
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Key PointsMedical Artificial Intelligence and Human ValuesAs large language models and other artificial intelligence models are used more in medicine, ethical dilemmas can arise depending on how the model was trained. A user must understand how human decisions and values can shape model outputs. Medical decision analysis offers lessons on measuring human values.A large language model will respond differently depending on the exact way a query is worded and how the model was directed by its makers and users. Caution is advised when considering the use of model output in decision making.
ISSN:0028-4793
1533-4406
1533-4406
DOI:10.1056/NEJMra2214183