2023Global SocietyOpen access

The Attributability of Combatant Status to Military AI Technologies under International Humanitarian Law

Mustafa Can Sati

Open full text 9 citations

Abstract

The concepts of means of warfare and combatants are not comparable or on the same scale in IHL. Yet the human-like performances of AI technologies, such as independent decision-making, may blur the line between these two concepts. This may also lead one to compare the technology with a human combatant rather than with other means of warfare. In this context, this paper questions the attributability of combatant status to military AI technologies by concentrating on the scope of the combatant concept. Contrary to some existing studies that found combatant status insufficient for machines based on ethics or behavioural human-machine differences, this study examines why combatant status is unsuitable for military AI technologies from a legal conceptual perspective, even in their most intelligent and independent forms by visiting terms—membership to armed forces, armed forces and prisoners of war (POW)—that are relevant to disclose the scope of the term combatant.

About this research paper

What this paper is about

The concepts of means of warfare and combatants are not comparable or on the same scale in IHL. Yet the human-like performances of AI technologies, such as independent decision-making, may blur the line between these two concepts. This may also lead one to compare the technology with a human combatant rather than with other means of warfare. In this context, this paper questions the attributability of combatant status to military AI technologies by concentrating on the scope of the combatant concept. Contrary to some existing studies that found combatant status insufficient for machines based on ethics or behavioural human-machine differences, this study examines why combatant status is unsuitable for military AI technologies from a legal conceptual perspective, even in their most intelligent and independent forms by visiting terms—membership to armed forces, armed forces and prisoners of war (POW)—that are relevant to disclose the scope of the term combatant.

Why it matters

OpenAlex reports 9 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

The concepts of means of warfare and combatants are not comparable or on the same scale in IHL. Yet the human-like performances of AI technologies, such as independent decision-making, may blur the line between these two concepts. This may also lead one to compare the technology with a human combatant rather than with other means of warfare. In this context, this paper questions the attributability of combatant status to military AI technologies by concentrating on the scope of the combatant concept. Contrary to some existing studies that found combatant status insufficient for machines based on ethics or behavioural human-machine differences, this study examines why combatant status is unsuitable for military AI technologies from a legal conceptual perspective, even in their most intelligent and independent forms by visiting terms—membership to armed forces, armed forces and prisoners of war (POW)—that are relevant to disclose the scope of the term combatant.

Key concepts: Combatant, Scope (computer science), International humanitarian law, Context (archaeology), Law, Engineering ethics, Perspective (graphical), Political science

Related papers

Back to paper searchBrowse research topicsOriginal source
The Attributability of Combatant Status to Military AI Technologies under International Humanitarian Law — Research Paper | ScholarLens