To prevent automated spam submissions leave this field empty.
The session was co-sponsored by two interest groups: Nonproliferation, Arms Control and Disarmament, and Space Law, and the. Lieber Society on the Law of Armed Conflict.
The moderator, Antonia Handler Chayes of the Fletcher School of International Affairs at Tufts University, opened by providing a couple of examples of the success and failure of armed control regimes and asking, in view of new technologies, what rules should apply, how they might develop, and how they can be implemented. Each speaker was then asked to comment on a certain aspect/category of the new technologies.
Steven Hill of the North Atlantic Treaty Organization focused on cyber activities. Arms control is an important issue for NATO, and in terms of cyberspace, like many experts in the field, NATO emphasized that international law remains applicable. NATO has seen a 60% increase in cyber incidents against its networks in 2016; nevertheless, NATO will act with restraint. One major question is whether a cyber attack an armed attack for the purpose of Article 5 of the North Atlantic Treaty. It is possible for Article 5 to be triggered in situations of cyber attack, and the determination should be made on a case-by-case basis, resorting to the classic armed attack definitions. Academic efforts, such as the Tallinn Manual, are of particular importance here. Further, this field needs more states, especially those with advance capabilities, to share their views on the applicability of international law. While it may not be necessary to develop new norms/conventions, efforts need to be made to clarify the rules applicable to cyberspace.
Tamara Patton of the Stockholm International Peace Research Institute discussed using new technologies to control one of the "old weapons" – nuclear weapons – as well as using lessons from the nuclear weapons domain to consider the challenges of new technologies. Firstly, the need for fostering public engagement was emphasized. Such engagement fueled the efforts of regulating nuclear weapons, which further contributed to prohibition of proliferation. Secondly, the experience with nuclear weapons might help guide the conversation of what type of transparency is desirable for different new technologies. It and other armed control regimes could also help with the exploration of verification means for new technologies. Lastly, references were also made to simulation software and the virtual environment and how they may stimulate ideas regarding the proper manner of arms control.
Sophie-Charlotte Fischer of the Center for Security Studies, Swiss Federal Institute in Zurich, analyzed issues concerning autonomous weapon systems (AWS). Currently, there does not seem to be consensus on the definition of such a term, or how such systems may look. Autonomy should be understood as a spectrum, and AWS do not require human input to select and engage targets of attack. On the one hand AWS have many attractive features, in terms of the efficiency, scope, and speed of attack, and of survivability. On the other hand, AWS clearly pose great challenges to the existing legal regimes, especially international humanitarian law. In order to fully comply with IHL norms (if that is in fact possible), such systems need to have knowledge of the broader context of each conflict and the specific conditions attached to each attack. It would also be necessary for them to be able to distinguish between combatants and non-combatants (which is especially difficult in asymmetric warfare) and to properly evaluate proportionality. AWS would also challenge the existing concepts of individual responsibility, and the risk of system failure is another issue that current regimes might not be able to address. Even against this background, the hardline position of banning AWS altogether is not necessary desirable or feasible, especially in view of the involvement of private actors in development and the speed at which relevant technologies are developing. Soft law measures may be considered at this stage to address the various concerns.
The last speaker, Paul Scharre of the Center for a New American Security, provided a historical perspective to the discussion. While there has been examples of success, restraint by states when faced with the development of new weapon technologies has proven to be difficult. If new weapons/technologies are to be banned, it is important to have clear focal points, and what can/cannot be used must be specified in order for the bans to be effective. It is easier to advocate banning weapons which cause superfluous harm. For AWS and artificial intelligence, it is harder, as regulating/banning them requires new decision-making models beyond the traditional regimes of arms control.
From the statements made by the speakers and the audience members who posed questions, it seemed to be generally agreed that banning new technologies/weapons would not be the answer. The task then becomes one of rethinking the current legal framework, examining the applicability of existing norms, and exploring the possible technological developments with caution and awareness of the relevant rules.
Pei-Lun Tsai is Post-Doctoral Fellow and Adjunct Assistant Professor at National Chengchi University, Taiwan.