This publication considers lethal autonomous weapon systems, approaching the issue from five different perspectives. It has been published ahead of the first meeting of the Group of Governmental Experts of the High Contracting Parties to the Convention on Certain Conventional Weapons mandated to examine issues related to emerging technologies in the area of lethal autonomous weapon systems in the context of the objectives and purposes of the Convention. The United Nations Office for Disarmament Affairs Occasional Papers are a series of ad hoc publications featuring, in edited form, papers or statements made at meetings, symposiums, seminars, workshops or lectures that deal with topical issues in the field of arms limitation, disarmament and international security.

Autonomous weapons systems seem to be on the path to becoming accepted technologies of warfare. The weaponization of artificial intelligence raises questions about whether human beings will maintain control of the use of force. The notion of meaningful human control has become a focus of international debate on lethal autonomous weapons systems among members of the United Nations: many states have diverging ideas about various complex forms of human-machine interaction and the point at which human control stops being meaningful. In Autonomous Weapons Systems and International Norms Ingvild Bode and Hendrik Huelss present an innovative study of how testing, developing, and using weapons systems with autonomous features shapes ethical and legal norms, and how standards manifest and change in practice. Autonomous weapons systems are not a matter for the distant future – some autonomous features, such as in air defence systems, have been in use for decades. They have already incrementally changed use-of-force norms by setting emerging standards for what counts as meaningful human control. As UN discussions drag on with minimal progress, the trend towards autonomizing weapons systems continues. A thought-provoking and urgent book, Autonomous Weapons Systems and International Norms provides an in-depth analysis of the normative repercussions of weaponizing artificial intelligence.

This open access book introduces the reader to the foundations of AI and ethics. It discusses issues of trust, responsibility, liability, privacy and risk. It focuses on the interaction between people and the AI systems and Robotics they use. Designed to be accessible for a broad audience, reading this book does not require prerequisite technical, legal or philosophical expertise. Throughout, the authors use examples to illustrate the issues at hand and conclude the book with a discussion on the application areas of AI and Robotics, in particular autonomous vehicles, automatic weapon systems and biased algorithms. A list of questions and further readings is also included for students willing to explore the topic further.

Winner of the 2019 William E. Colby Award "The book I had been waiting for. I can't recommend it highly enough." —Bill Gates The era of autonomous weapons has arrived. Today around the globe, at least thirty nations have weapons that can search for and destroy enemy targets all on their own. Paul Scharre, a leading expert in next-generation warfare, describes these and other high tech weapons systems—from Israel’s Harpy drone to the American submarine-hunting robot ship Sea Hunter—and examines the legal and ethical issues surrounding their use. “A smart primer to what’s to come in warfare” (Bruce Schneier), Army of None engages military history, global policy, and cutting-edge science to explore the implications of giving weapons the freedom to make life and death decisions. A former soldier himself, Scharre argues that we must embrace technology where it can make war more precise and humane, but when the choice is life or death, there is no replacement for the human heart.

"Because of the increasing use of Unmanned Aerial Vehicles (UAVs, also commonly known as drones) in various military and para-military (i.e., CIA) settings, there has been increasing debate in the international community as to whether it is morally and ethically permissible to allow robots (flying or otherwise) the ability to decide when and where to take human life. In addition, there has been intense debate as to the legal aspects, particularly from a humanitarian law framework. In response to this growing international debate, the United States government released the Department of Defense (DoD) 3000.09 Directive (2011), which sets a policy for if and when autonomous weapons would be used in US military and para-military engagements. This US policy asserts that only "human-supervised autonomous weapon systems may be used to select and engage targets, with the exception of selecting humans as targets, for local defense ...". This statement implies that outside of defensive applications, autonomous weapons will not be allowed to independently select and then fire upon targets without explicit approval from a human supervising the autonomous weapon system. Such a control architecture is known as human supervisory control, where a human remotely supervises an automated system (Sheridan 1992). The defense caveat in this policy is needed because the United States currently uses highly automated systems for defensive purposes, e.g., Counter Rocket, Artillery, and Mortar (C-RAM) systems and Patriot anti-missile missiles. Due to the time-critical nature of such environments (e.g., soldiers sleeping in barracks within easy reach of insurgent shoulder-launched missiles), these automated defensive systems cannot rely upon a human supervisor for permission because of the short engagement times and the inherent human neuromuscular lag which means that even if a person is paying attention, there is approximately a half-second delay in hitting a firing button, which can mean the difference for life and death for the soldiers in the barracks. So as of now, no US UAV (or any robot) will be able to launch any kind of weapon in an offensive environment without human direction and approval. However, the 3000.09 Directive does contain a clause that allows for this possibility in the future. This caveat states that the development of a weapon system that independently decides to launch a weapon is possible but first must be approved by the Under Secretary of Defense for Policy (USD(P)); the Under Secretary of Defense for Acquisition, Technology, and Logistics (USD(AT&L)); and the Chairman of the Joint Chiefs of Staff. Not all stakeholders are happy with this policy that leaves the door open for what used to be considered science fiction. Many opponents of such uses of technologies call for either an outright ban on autonomous weaponized systems, or in some cases, autonomous systems in general (Human Rights Watch 2013, Future of Life Institute 2015, Chairperson of the Informal Meeting of Experts 2016). Such groups take the position that weapons systems should always be under "meaningful human control," but do not give a precise definition of what this means. One issue in this debate that often is overlooked is that autonomy is not a discrete state, rather it is a continuum, and various weapons with different levels of autonomy have been in the US inventory for some time. Because of these ambiguities, it is often hard to draw the line between automated and autonomous systems. Present-day UAVs use the very same guidance, navigation and control technology flown on commercial aircraft. Tomahawk missiles, which have been in the US inventory for more than 30 years, are highly automated weapons with accuracies of less than a meter. These offensive missiles can navigate by themselves with no GPS, thus exhibiting some autonomy by today's definitions. Global Hawk UAVs can find their way home and land on their own without any human intervention in the case of a communication failure. The growth of the civilian UAV market is also a critical consideration in the debate as to whether these technologies should be banned outright. There is a $144.38B industry emerging for the commercial use of drones in agricultural settings, cargo delivery, first response, commercial photography, and the entertainment industry (Adroit Market Research 2019) More than $100 billion has been spent on driverless car development (Eisenstein 2018) in the past 10 years and the autonomy used in driverless cars mirrors that inside autonomous weapons. So, it is an important distinction that UAVs are simply the platform for weapon delivery (autonomous or conventional), and that autonomous systems have many peaceful and commercial uses independent of military applications"--

The deployment of Autonomous Weapons gives rise to ongoing debate in society and at the United Nations, in the context of the Convention on Certain Conventional Weapons. Yet little empirical research has been done on this topic. This volume fills that gap by offering an empirical study based on military personnel and civilians working at the Dutch Ministry of Defence. It yields insight into how Autonomous Weapons are perceived by the military and general public; and which moral values are considered important in relation to their deployment. The research approach used is the Value-Sensitive Design (VSD) method that allows for the consideration of human values throughout the design process of technology. The outcome indicates that military personnel and civilians attribute more agency (the capacity to think and plan) to an Autonomous Weapon than to a Human Operated Drone. In addition, it is clear that common ground exists between military and societal groups in their perception of the values of human dignity and anxiety. These two values arise often in the discourse, and addressing them is essential when considering the ethics of the deployment of Autonomous Weapons. The text of this volume is also offered in parallel French and German translation.

"This 50-page report outlines concerns about these fully autonomous weapons, which would inherently lack human qualities that provide legal and non-legal checks on the killing of civilians. In addition, the obstacles to holding anyone accountable for harm caused by the weapons would weaken the law's power to deter future violations"--Publisher's website.

The main theme of this volume of the Yearbook of International Humanitarian Law is weapons law. In several chapters, how International Humanitarian Law (IHL) copes with old and new weapons as well as political developments in regard to military technology is discussed, while in two chapters the significance of non- or less-lethal weapons in peace-keeping and law enforcement operations as well as the legality of lethal autonomous weapon systems under IHL are analysed. Moreover, the volume describes the current status of nuclear deterrence under international law. Another layer is added by examining how IHL influences the programming of automatic target recognition systems using artificial intelligence. The second part of the book contains a historic perspective on the roots of IHL in Europe, which can be traced back to the ninth century, as well as a Year in Review describing the most important events and legal developments in the area of IHL that took place in 2018. The Yearbook of International Humanitarian Law is the world’s only annual publication devoted to the study of the laws governing armed conflict. It provides a truly international forum for high-quality, peer-reviewed academic articles focusing on this crucial branch of international law. Distinguished by contemporary relevance, the Yearbook of International Humanitarian Law bridges the gap between theory and practice and serves as a useful reference tool for scholars, practitioners, military personnel, civil servants, diplomats, human rights workers and students.

The DSB Task Force on the Role of Autonomy in DoD Systems was asked to study relevant technologies, ongoing research, and the current autonomy-relevant plans of the Military Services, to assist the DoD in identifying new opportunities to more aggressively use autonomy in military missions, to anticipate vulnerabilities, and to make recommendations for overcoming operational difficulties and systemic barriers to realizing the full potential of autonomous systems. The Task Force has concluded that, while currently fielded unmanned systems are making positive contributions across DoD operations, autonomy technology is being underutilized as a result of material obstacles within the Department that are inhibiting the broad acceptance of autonomy and its ability to more fully realize the benefits of unmanned systems. Overall, the Task Force found that unmanned systems are making a significant, positive impact on DoD objectives worldwide. However, the true value of these systems is not to provide a direct human replacement, but rather to extend and complement human capability by providing potentially unlimited persistent capabilities, reducing human exposure to life threatening tasks, and with proper design, reducing the high cognitive load currently placed on operators/supervisors.Unmanned systems are proving to have a significant impact on warfare worldwide. The true value of these systems is not to provide a direct human replacement, but rather to extend and complement human capability in a number of ways. These systems extend human reach by providing potentially unlimited persistent capabilities without degradation due to fatigue or lack of attention. Unmanned systems offer the warfighter more options and flexibility to access hazardous environments, work at small scales, or react at speeds and scales beyond human capability. With proper design of bounded autonomous capabilities, unmanned systems can also reduce the high cognitive load currently placed on operators/supervisors. Moreover, increased autonomy can enable humans to delegate those tasks that are more effectively done by computer, including synchronizing activities between multiple unmanned systems, software agents and warfighters--thus freeing humans to focus on more complex decision making.1.0 Executive Summary * 1.1. Misperceptions about Autonomy are Limiting its Adoption * 1.2. Create an Autonomous Systems Reference Framework to Replace "Levels of Autonomy" * 1.3. Technical Challenges Remain, Some Proven Autonomy Capability Underutilized * 1.4. Autonomous Systems Pose Unique Acquisition Challenges * 1.5. Avoid Capability Surprise by Anticipating Adversary Use of Autonomous Systems * 2.0 Operational Benefits of Autonomy * 2.1. Unmanned Aerial Vehicles * 2.2. Unmanned Ground Systems * 2.3. Unmanned Maritime Vehicles * 2.4. Unmanned Space Systems * 2.5. Conclusion * 3.0 Technical Issues of Autonomy * 3.1. Motivation: What Makes Autonomy Hard * 3.2. Defining Levels of Autonomy is Not Useful * 3.3. Autonomous System Reference Framework * 3.4. Needed Technology Development * 3.5. Technical Recommendations * 4.0 Acquisition Issues of Autonomy * 4.1. Requirements and Development * 4.2. Test and Evaluation * 4.3. Transition to Operational Deployment * 5.0 Capability Surprise in Autonomy Technology * 5.1. Overview of Global Unmanned Systems * 5.2. Unmanned Symmetric Adversary Scenarios * 5.3. Value for Asymmetric Adversaries * 5.4. External Vulnerabilities * 5.5. Self-Imposed Vulnerabilities * 5.6. Recommendations . * Appendix A--Details of Operational Benefits by Domain * A.1. Aerial Systems Strategy * A.2. Maritime Systems * A.3. Ground Systems * A.4. Space Systems * Appendix B--Bibliography * Appendix C--Task Force Terms of Reference * Appendix D--Task Force Membership * Appendix E--Task Force Briefings * Appendix F--Glossary

Expounding on the results of the author's work with the US Army Research Office, DARPA, the Office of Naval Research, and various defense industry contractors, Governing Lethal Behavior in Autonomous Robots explores how to produce an "artificial conscience" in a new class of robots, humane-oids, which are robots that can potentially perform more et