… to turn to your first questions, and these are, of course, kind of preliminary remarks, I would say that legal reviews by themselves are not the only tool to ensure compliance with IHL, and I agree with my colleague from Ecuador, who made the same point that there are so many implementing measures for [IHL]. And we have to remember that IHL is in itself very much a preventive framework. And that compliance with IHL is always something that should be ensured before weapons are actually used in a specific situation. So that brings me actually to the second point, where you talk about, where you asked the question about fragmentation of norms, and there may be differences of opinion about the legality or illegality of the use of certain weapons with autonomous functions. I think we must be very clear that there’s only one set of [IHL], there is no [IHL] as understood by the military or [IHL] as understood by humanitarian organisation. It’s just that one body of [IHL], which must be interpreted, to get back to one of your questions from yesterday, by the international law interpretation methods that are there in international law. And this is for states, and other stakeholders can help in that respect. And I think there are many different ways we can avoid the risk that there are misunderstandings of different interpretations of how weapons should be used, and by different states. And the idea of vehicle, I’m not sure whether it’s an ideal vehicle, but I think that one thing that we must continue to do is to discuss these issues with each other. And I like to echo one of the points that was made before that it’s important to look at the technical aspects — what do these things actually do. And I think just listening to the contribution from the University of Cambridge made me realise how little I really understand from all these technical aspects, and I’m not sure whether, as a lawyer, I’m alone in this room in that respect, but I do think there’s much more room in that respect. And I would like to use this opportunity to encourage everybody to bring tech people to the GGE LAWS and other meetings that we have amongst each other, to really grasp the understanding of what AI really means, or what autonomous functions really mean. So that brings me to the last question that you asked. To carry out weapon reviews, do we need a definition of lethal or any type of [AWS]? As a lawyer, I would always like to return to the law for questions like this. And the law says in Article 36, that we must assess the means and methods of warfare. So that means that we are not looking only at the weapon itself; in fact, we have to look at the way in which it is used and the method of warfare. So I think that already makes the point that a definition, a technical definition, in itself is not necessarily required to actually make an assessment of the use of a specific means or method of warfare in a specific context. So I think that we can still really look at the legality of means and methods of warfare, including autonomous means or methods of warfare with autonomous functions. So, and the last point maybe that I could make in this respect, is to make that assessment, there’s actually — it’s a bit older now — but there’s a guide, which was published by the ICRC on Article 36 legal reviews mechanism. So in case there are still uncertainties of how these legal reviews could be conducted, I would suggest that maybe that guide by the ICRC is a useful tool or a first step for guidance for how these legal reviews can be conducted.
Statement by the Netherlands under agenda item 5, topic 5 (9 March 2023, 2nd intervention, transcript)
