Cybertrust: From Explainable to Actionable and Interpretable Artificial Intelligence

We argue that artificial intelligence (AI) systems should be designed with features that build trust by bringing decision-analytic perspectives into AI. Actionable and interpretable AI will incorporate explicit quantifications and visualizations of user confidence in AI recommendations.

Saved in:

Bibliographic Details
Published in:	Computer (Long Beach, Calif.) Vol. 53; no. 9; pp. 91 - 96
Main Authors:	Linkov, Igor, Galaitsi, Stephanie, Trump, Benjamin D., Keisler, Jeffrey M., Kott, Alexander
Format:	Journal Article
Language:	English
Published:	New York IEEE 01-09-2020 The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Subjects:	Artificial intelligence Cognition Decision analysis Decision making Heart Machine learning Prediction algorithms Stakeholders
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	We argue that artificial intelligence (AI) systems should be designed with features that build trust by bringing decision-analytic perspectives into AI. Actionable and interpretable AI will incorporate explicit quantifications and visualizations of user confidence in AI recommendations.
ISSN:	0018-9162 1558-0814
DOI:	10.1109/MC.2020.2993623