Unpacking the responsibility gap(s) arising from AI applications in weapon technologies

On Wednesday 19 June T.M.C. Asser Instituut will be hosting an event entitled “Unpacking the responsibility gap(s) arising from AI applications in weapon technologies”. This symposium will address both the potential responsibility gaps in individual criminal responsibility for war crimes and State responsibility that may arise in relation to the implementation of artificial intelligence (AI) in weapon systems.

Responsibility under international criminal law and the doctrine of State responsibility serve different, but complementary, functions in international law. International criminal law seeks to hold individual perpetrators accountable for the commission of grave international crimes, whilst State responsibility pursues accountability at the State level for internationally wrongful acts. The advent of AI in weapon systems raises challenges for ensuring accountability under international law in ways that are not yet fully understood in academic and diplomatic circles. The discourse at the United Nations Group of Governmental Experts (GGE) on Lethal Autonomous Weapon Systems (LAWS) highlights this lack of comprehension, with delegations often referring to the need ensure accountability and address responsibility gaps, without engaging with this issue in an in-depth manner.

The purpose of this symposium is to identify the main issues at stake in ascribing responsibility, at both the State and individual level, for violations of international law committed with autonomous weapons. It is hoped that this will provide greater clarity around the potential responsibility gap(s) under international law to guide future discussions on the emerging use of artificial intelligence in weapon systems. 

The speakers

• Alice Beck (Project Officer Autonomous Weapons, PAX)
• Dr Marta Bo (Researcher, Asser Institute; Graduate Institute, Geneva)
• Dr Berenice Boutin (Researcher, Asser Institute)

The programme

Alice Beck will present PAX’s latest report ‘The State of AI: Artificial Intelligence, the Military and Increasingly Autonomous Weapons’, published as part of the Reprogramming War project. The report analyses developments of AI in seven key countries, providing an insight into AI military strategies as well as AI-related military projects and weapon systems. ‘State of AI’ therefore illustrates the trend of increasing autonomy in weapon systems. The aim of this presentation is to give a background of the current context in order to inform the legal discussions on responsibility. 

Dr. Marta Bo will assess the challenges posed by the introduction of AI in weapons to the ascription of criminal responsibility for war crimes. She will highlight how discussions at diplomatic and academic levels have fallen short in engaging with the issue of criminal responsibility and attempt to re-situate this legal debate in order to foster robust analysis of the impact of autonomy in weapons on the ascription of criminal responsibility.

Dr Berenice Boutin will discuss the issue of State responsibility in relation to violations of international law involving military AI technologies. At the stage of deployment, when AI technologies are used in the battlefield, the responsibility of the State can be engaged if it is demonstrated that the wrongful conduct in question is attributable to the State. However, existing rules of State responsibility implicitly rely on an element of human agency. The question is therefore whether and how the increasing autonomous capacities of certain AI systems could affect the operation of attribution of conduct. Besides, the responsibility of States can also be invoked with regards to wrongful conduct at the stage of the development or acquisition of AI. Indeed, AI must be designed and developed in full compliance with existing applicable international norms, and failure to do so can engage the responsibility of a State.