On the Meaning of AI Safety

Abstract

There is a growing urgency and global demand to address Artificial Intelligence (AI) safety. However, swift and sustained progress is unlikely to emerge without a shared understanding of what it means for the use of AI to be safe. This paper advances a comprehensive definition of AI safety and explores the fundamental concepts underpinning this definition. The aim is to contribute to a meaningful and inclusive discussion and further the public discourse on AI safety.

Metadata

Item Type:	Monograph
Authors/Creators:	Habli, Ibrahim https://orcid.org/0000-0003-2736-8238
Keywords:	Artificial Intelligence,Safety,Risk,Harm
Dates:	Submitted: 17 March 2024 Published: 8 April 2025
Institution:	The University of York
Academic Units:	The University of York > Faculty of Sciences (York) > Computer Science (York)
Depositing User:	Pure (York)
Date Deposited:	26 Oct 2023 11:30
Last Modified:	22 Apr 2025 23:10
Status:	Published
Open Archives Initiative ID (OAI ID):	oai:eprints.whiterose.ac.uk:204545