A new treaty would clarify and strengthen existing international humanitarian law. It would establish clear international rules to address the specific problem of weapons systems operating outside meaningful human control. In this way, the instrument would fill the legal gap identified by the Martens Clause, help resolve interpretation disputes, promote consistency in interpretation and implementation, and facilitate compliance and enforcement. [13] In the case of fully autonomous weapons, it would be even more difficult to weigh the proportionality of an attack. The proportionality test must determine whether the civilian harm expected outweighs the military benefits expected in a rapidly changing environment. Assessing the proportionality of an attack requires more than a quantitative calculation. Commanders apply human judgment, shaped by legal and moral norms and personal experience, to the specific situation. It is doubtful that the human judgment required to assess proportionality can ever be replicated in a machine. Moreover, the robots could not be programmed in advance to deal with the infinite number of unexpected situations they might encounter on the battlefield. [3] The use of fully autonomous weapons also risks creating a serious accountability gap. [4] International humanitarian law requires that individuals be held legally responsible for war crimes and grave breaches of the Geneva Conventions. Military commanders or operators could be convicted of using a fully autonomous weapon with intent to commit a crime. However, it would be legally difficult and arguably unfair to hold an operator accountable for the unpredictable actions of an autonomous robot.

The concept of meaningful human control is crucial to the new treaty, as the moral, legal, and liability issues associated with fully autonomous weapons are largely due to the lack of such control. [19] Recognizing these risks, most states have adopted the principle that individuals must play a role in the use of force. [20] Although they have used different terminology, many states and experts prefer the term “meaningful human control”. “Control” is stronger than alternatives such as “intervention” and “judgment” and is broad enough to encompass both; It is also a well-known concept in international law. [21] “Reasonable” ensures that control reaches a meaningful level. [22] Other organizations, however, place the autonomous weapon system standard in a higher position. The UK Ministry of Defence defines autonomous weapons systems as “systems capable of understanding intent and direction at a higher level. From this understanding and perception of its environment, such a system is able to take appropriate measures to cause a desired state. It is able to decide on a course of action from a range of alternatives without relying on human oversight and control – however, such human commitment to the system may still be present.

While the overall activity of an autonomous unmanned aircraft is predictable, individual actions may not be predictable. [6] Lethal autonomous weapons systems (LAWS), or weapons designed to independently select and engage targets without the need for manual human control, could enable military operations in disrupted or denied communication environments where conventional systems may not be able to function. LAWS are not yet widely used. However, with advances in technology – particularly artificial intelligence (AI) – more countries may consider developing and enforcing laws. This could have potential implications for congressional oversight, defense investments, military concepts of operations, treaty design, and the future of war. A 2021 report from the U.S. Congressional Research Service states that “there are no national or international legal prohibitions on expanding the use of LAWS,” although it acknowledges ongoing discussions under the United Nations Convention on Certain Conventional Weapons (CCW). [37] The proposal for a new Treaty recommends a broad scope. The treaty should apply to any weapon system that selects and attacks targets based on sensor processing rather than human intervention. [18] The scope is intended to ensure that all systems in this category, whether present or future, are assessed and that problematic systems do not escape regulation. However, the prohibitions and restrictions described in detail below are forward-looking and focus on fully autonomous weapons. Scientists like Peter Asaro and Mark Gubrud are trying to lower the threshold and evaluate more weapons systems as autonomous.

They believe that any weapon system capable of unleashing lethal force without the operation, decision or confirmation of a human superior can be considered autonomous. According to Gubrud, a weapon system that operates partially or completely without human intervention is considered autonomous. He argues that a weapons system does not need to be able to make decisions completely on its own to be called autonomous. Instead, it should be treated as autonomous as long as it is actively involved in one or more parts of the “preparation process,” from target finding to final firing. [4] [5] Some recent examples of LAW are automated “hardkill” active protection systems, such as: radar-controlled CIWS systems for the defense of ships in service since the 1970s (e.g. the American CIWS phalanx). These systems can autonomously identify and attack missiles, missiles, artillery fire, aircraft and surface ships in the opposite direction according to criteria set by the human operator. Similar systems exist for tanks, such as the Russian Arena, the Israeli Trophy and the German AMAP-ADS. Different types of fixed guard weapons that can fire at people and vehicles are used in South Korea and Israel.

Many missile defense systems, such as the Iron Dome, also have autonomous targeting capabilities. Fully autonomous weapons pose many other threats that go far beyond concerns about respect for international humanitarian law. For many, delegating life-and-death decisions to machines would cross a moral red line. [9] The use of fully autonomous weapons, including in law enforcement operations, would violate the rights to life, reparation, and dignity. [10] The development and production of these machines could trigger an arms race, and the systems could spread to irresponsible states and non-state armed groups. [11] Even if new technologies could solve some of the problems of international humanitarian law discussed above, they would not solve many of these other problems. [10] See generally Human Rights Watch and IHRC, Shaking the Foundations: The Human Rights Implications of Killer Robots (May 2014), www.hrw.org/sites/default/files/reports/arms0514_ForUpload_0.pdf (accessed May 20, 2020). See also Heyns, “Lethal Autonomous Robotics”, pp. 6 (on the right to life: “The introduction of such powerful but controversial new weapons systems may again threaten the right to life”), 15 (on the right to reparation: “If the nature of a weapon makes it impossible to be responsible for its consequences, its use should be considered unethical and illegal as a heinous weapon”) and 20 (on dignity: “There is a general concern that allowing [fully autonomous weapons] to kill people may denigrate the value of life itself.”) Being “autonomous” has different meanings depending on the fields of study. In engineering, it can refer to the machine`s ability to operate without human intervention. In philosophy, it can refer to an individual who is morally independent. In political science, it can refer to the capacity for self-government of a domain.

In terms of military weapons development, the identification of a weapon as autonomous is not as clear as in other areas. [1] The specific norm associated with the concept of autonomy can vary considerably between different scientists, nations and organizations. Terms like “slaughter robots” and “killer robots” are reminiscent of sci-fi movies like Terminator, which features a confident, human-like assassin robot. This fuels the assumption that lethal autonomous weapons are of the distant future. After six years of discussions on certain conventional weapons, States should actively consider the elements of a new treaty and conduct negotiations on its implementation. Theoretically, the negotiations could lead to a new CCW protocol, but some states have used the CCW consensus rules to block progress. It is therefore time to consider another forum. States could initiate an independent process, such as the one used to create the Mine Ban Treaty or the Convention on Cluster Munitions, or they could adopt a treaty under the auspices of the United Nations General Assembly, as was the case with the Arms Trade Treaty and the Treaty on the Prohibition of Nuclear Weapons. Ultimately, states should follow the most effective path to the most effective treaty that prevents the dangers posed by fully autonomous weapons. States parties to the Convention on Certain Conventional Weapons (CCW) have held eight in-depth meetings on lethal autonomous weapons systems since 2014.

They examined the broad challenges of systems and recognized the importance of maintaining human control over the use of force. However, progress towards an appropriate multilateral solution has been slow. If states do not move quickly from abstract talks to treaty negotiations, the development of technology will outpace international diplomacy. The Holy See has repeatedly called on the international community to ban the use of LAW.