In this paper, we distinguish between different sorts of assessments of algorithmic systems, describe our process of assessing such systems for ethical risk, and share some key challenges and lessons for future algorithm assessments and audits. Given the distinctive nature and function of a third-party audit, and the uncertain and shifting regulatory landscape, we suggest that second-party assessments are currently the primary mechanisms for analyzing the social impacts of systems that incorporate artificial intelligence. We then discuss two kinds of as-sessments: (...) an ethical risk assessment and a narrower, technical algo-rithmic bias assessment. We explain how the two assessments depend on each other, highlight the importance of situating the algorithm within its particular socio-technical context, and discuss a number of lessons and challenges for algorithm assessments and, potentially, for algorithm audits. The discussion builds on our team’s experience of advising and conducting ethical risk assessments for clients across dif-ferent industries in the last four years. Our main goal is to reflect on the key factors that are potentially ethically relevant in the use of algo-rithms, and draw lessons for the nascent algorithm assessment and audit industry, in the hope of helping all parties minimize the risk of harm from their use. (shrink)

It is common to think of warfare as a setting in which emotion can lead combatants to engage in unethical behavior. On this view, it is natural to conceptualize the aim of military ethics training as quelling the influence of emotion in combat in order to reduce the risk that military personnel are vulnerable to its influence. Recent research, however, indicates that what is called “emotion processing” is connected in important ways with moral judgment and behavior. In this view, acting (...) ethically is not simply a matter of subordinating emotion to reason, but of responding with appropriate rather than inappropriate emotion. In this respect, such research challenges the powerful “rationalist” account of moral judgment and behavior. More fundamentally, it calls into question the distinction between reason and emotion as a way to understand our moral sensibilities. This article describes this research and discusses work on naturalistic decision-making and neuroscience that provide support for its insights. It then suggests ways in which such research is consistent with Aristotelian notions of character and virtue. Finally, it identifies ways in which some military training implicitly draws on lessons from this body of research, and suggests how it might do so more deliberately. (shrink)

The trend toward the “humanization” of international law reflects a greater emphasis on individuals rather than simply states as objects of concern. The advance of human rights law (HRL) has been an important impetus for this trend. Some observers suggest that humanization can be furthered even more by applying HRL rather than international humanitarian law (IHL) to hostilities between states and nonstate armed groups, unless a state explicitly declares that it is engaged in an armed conflict. This essay argues, however, (...) that a court should not defer to a state's characterization of hostilities, but should base its analysis on whether hostilities meet the criteria for an armed conflict. Applying HRL to hostilities that effectively are an armed conflict but not acknowledged as such risks diluting the legitimacy and normative force of HRL. On the one hand, if a court applies conventional stringent HRL standards, this body of law may be seen as unrealistic and is likely to be ignored. On the other hand, a court that adapts HRL standards to armed conflict may need to take a consequentialist approach at odds with HRL's deontological foundations. Clearly differentiating between HRL and IHL may thus best promote the humanization of warfare. (shrink)

This volume examines the ethical issues that arise as a result of national security intelligence collection and analysis. Powerful new technologies enable the collection, communication, and analysis of national security data on an unprecedented scale. Data collection now plays a central role in intelligence practice, yet this development raises a host of ethical and national security problems, such as: privacy; autonomy; threats to national security and democracy by foreign states; and accountability for liberal democracies. This volume provides a comprehensive set (...) of in-depth ethical analyses of these problems by combining contributions from both ethics scholars and intelligence practitioners. It provides the reader with a practical understanding of relevant operations, the issues that they raise, and analysis of how responses to these issues can be informed by a commitment to liberal democratic values. This combination of perspectives is crucial in providing an informed appreciation of ethical challenges that is also grounded in the realities of the practice of intelligence. This book will be of great interest to all students of intelligence studies, ethics, security studies, foreign policy, and International Relations. (shrink)

ABSTRACTIn November 2005, U.S. Marine Sergeant Frank Wuterich fired on and killed five unarmed Iraqi men standing by a car near the site of an improvised explosive device explosion in Haditha...