- 03/05/2019
- Category: Commentaries
The use and development of Artificial Intelligence (AI) have greatly indicated significant technological progress for the future of the humankind. It has generated abundant opportunities to come out across many areas, which include labor, medicine, national security, transportation, safety, and so forth[1] – thus, creating more efficiencies and feasibilities on humans’ lives. Nevertheless, there have been numerous debates and dilemmas upon its evolvement due to the unneglectable potential challenges or risks that it might pose – among them is the potential use and development of LAWS, which are weapons that could independently select and attack people who are considered as threats to humanity as their “critical functions”, with little to no human intervention.[2] Recent developments of these weapons systems, which are exacerbated by the increasing military operational demands, have reduced or eliminated the potential of direct human control over the use of force in general.[3] They could be either fully autonomous, in which the absence of human intervention takes place during their launch – or semi-autonomous, which require human action to run their mission.[4] However, the main question relies upon whether or not their use and development could be protecting or threatening the humankind – especially by taking the hostile attitudes of the international community into account – mainly due to the potential ethical and legal challenges on the international humanitarian law (IHL) they might entail.
LAWS – or the so-called “killer robots” – are often regarded as being disrespectful to the requirements of jus in bello of IHL, which include the principles of distinction between civilians and combatants, proportionality between means and ends, and military necessity for the use of force to apply – not to mention the Martens Clause[5] on minimizing the potential of civilian casualties to occur. It is often argued by AI researchers and developers that the weapons possess the risk of targeting innocent civilians due to their potentials to be hacked or falling onto the hands of “irresponsible actors”[6], who could unlawfully leverage the weapons to attack non-combatants on the motive of pursuing their interests during armed conflicts. Hence, it is reasonable for them to have called for the ban of those weapons to the United Nations (UN) on the basis of unaccountability and disproportionality, which could be indicated by how nearly 3,100 individuals and 240 AI-related organizations have pledged to not involve themselves in the development of LAWS. Despite the fact that there have been no weapons that could specifically target humans, it is predicted by experts that it would be possible in just a few years from now.[7] In the worst case, it is claimed that global AI arms race could be ignited if countries begin to initiate substantial development of LAWS. [8] Meanwhile, from the ethical perspective, these “killer robots” might be considered as problematic in regards of the upholding of human dignity due to the consideration of humans as “objects” during warfare – life-death decisions are taken by robots, which are considered as lacking capacity to understand what it means to kill a human being.
On the other side of the spectra in terms of the lawfulness of LAWS, one might argue that LAWS could be more accurate and accountable in terms of decision-making process due to the precise collection of complex information – by also setting aside human feelings simultaneously, which would unquestionably lead to fairness and reduce more harmfulness to occur compared to human action.[9] Furthermore, it could also be argued that the accountable use of LAWS requires human intervention in the first place throughout the process of creation and testing, such as a team of engineers, military personnel, scientists, and so forth[10]; therefore, it is rather mistaken to claim that the absence of human intervention, including in fully autonomous LAWS, prevails. On the other hand, it is also claimed that the close-in weapon system (CIWS), which has a similar mechanism to how LAWS work – with missiles as the targets – has existed and been utilized for a rather long time.[11] Hence, the fear of the development of LAWS due to unlawful target killings, from both ethical and legal perspectives, is rather debatable due to the same logic applied to the implementation of CIWS.
It could be seen that the underlying ethical and legal principles of the development of LAWS have raised further question and dilemma of whether or not the development of those weapons would be legitimate to alleviate the potential of civilian casualties in armed conflicts. It actually depends on the systematic accountability of their use, from the creation to the execution. Despite being fully autonomous, human control still prevails, in which human responsibility takes more place throughout the practices than its “independent functions”. While they might be useful to some extent in terms of accuracy and preciseness due to the minimum human intervention, it is the responsible human control and setup that would determine the further justification of their deployment during conflicts.
Editor: Janitra Haryanto
Read another article written by Felice Valeria
[1] Dizikes, P. (2019). AI, the law, and our future. [online] MIT News. Available at: http://news.mit.edu/2019/first-ai-policy-congress-0118 [Accessed 15 Apr. 2019].
[2] International Committee of the Red Cross (2019). International Humanitarian Law and the Challenges of Contemporary Armed Conflicts. [online] Geneva: 32nd International Conference of the Red Cross and Red Descent, p.44. Available at: https://www.icrc.org/en/download/file/15061/32ic-report-on-ihl-and-challenges-of-armed-conflicts.pdf [Accessed 16 Apr. 2019].
[3] Ibid.,
[4] Evans, H. and Salmanowitz, N. (2019). Lethal Autonomous Weapons Systems: Recent Developments. [online] Lawfare. Available at: https://www.lawfareblog.com/lethal-autonomous-weapons-systems-recent-developments [Accessed 16 Apr. 2019].
[5] Casebook.icrc.org. (n.d.). Martens Clause | How does law protect in war?. [online] Available at: https://casebook.icrc.org/glossary/martens-clause [Accessed 16 Apr. 2019].
[6] Conn, A. (2018). The Risks Posed by Lethal Autonomous Weapons. [online] Future of Life Institute. Available at: https://futureoflife.org/2018/09/04/the-risks-posed-by-lethal-autonomous-weapons/?cn-reloaded=1 [Accessed 16 Apr. 2019].
[7] Ibid.,
[8] Ibid.,
[9] Petman, J. (2017). Autonomous Weapons Systems and International Humanitarian Law: ‘Out of the Loop’?. [ebook] Helsinki: Unigrafia Oy. Available at: https://um.fi/documents/35732/48132/autonomous_weapon_systems_an_international_humanitarian_law__out_of_the/c0fca818-3141-b690-0337-7cfccbed3013?t=1525645981157 [Accessed 16 Apr. 2019].
[11] Alton, L. (2019). The panic over lethal automated weapons systems is unfounded, and could be dangerous. [online] The Next Web. Available at: https://thenextweb.com/contributors/2018/11/24/the-panic-over-lethal-automated-weapons-systems-is-unfounded-and-could-be-dangerous/ [Accessed 16 Apr. 2019].
