Many ethics initiatives have stipulated sets of principles and standards for good technology development in the AI sector. However, several AI ethics researchers have pointed out a lack of practical realization of these principles. Following that, AI ethics underwent a practical turn, but without deviating from the principled approach. This paper proposes a complementary to the principled approach that is based on virtue ethics. It defines four “basic AI virtues”, namely justice, honesty, responsibility and care, all of which represent specific (...) motivational settings that constitute the very precondition for ethical decision making in the AI field. Moreover, it defines two “second-order AI virtues”, prudence and fortitude, that bolster achieving the basic virtues by helping with overcoming bounded ethicality or hidden psychological forces that can impair ethical decision making and that are hitherto disregarded in AI ethics. Lastly, the paper describes measures for successfully cultivating the mentioned virtues in organizations dealing with AI research and development. (shrink)

The Data Hazards framework (Zelenka, Di Cara, & Contributors, 2024) is intended to encourage thinking about the ethical implications of data science projects. It takes the form of community-designed data hazard labels, similar to warning labels on chemicals, that can encourage reflection and discussion on what ethical risks are associated with a project and how they can be mitigated. In this article, we explain how the Data Hazards framework can apply to neuroscience. We demonstrate how the hazard labels can be (...) applied to one of our own projects, on the computational modelling of postsynaptic mechanisms. Graphical Abstract. (shrink)

Due to the technology of machine learning, new breakthroughs are currently being achieved with constant regularity. By using machine learning techniques, computer applications can be developed and used to solve tasks that have hitherto been assumed not to be solvable by computers. If these achievements consider applications that collect and process personal data, this is typically perceived as a threat to information privacy. This paper aims to discuss applications from both fields of personality and image analysis. These applications are often (...) criticized by reference to the protection of privacy. This paper critically questions this approach. Instead of solely using the concept of privacy to address the risks of machine learning, it is increasingly necessary to consider and implement ethical anti-discrimination concepts, too. In many ways, informational privacy requires individual information control. However, not least because of machine learning technologies, information control has become obsolete. Hence, societies need stronger anti-discrimination tenets to counteract the risks of machine learning. (shrink)

Particular applications of Privacy by Design (PbD) have proven to be valuable tools to protect privacy in many technological applications. However, PbD is not as promising when applied to technologies used for surveillance. After specifying how surveillance and privacy are understood in this paper, I will highlight the shortcomings of PbD when applied to surveillance, using a web-scanning system for counter-terrorism purposes as an example. I then suggest reworking PbD into a different approach: the Minimum Harm by Design (MHbD) model. (...) MHbD differs from PbD principally in that it acknowledges that the potential harms of surveillance bear not only upon privacy but also values that define the very constitution of a society and its political character. MHbD aims to identify and systematise the different categories of such harms and links them to current theories on surveillance on the one hand and on possible design measures on the other. (shrink)

The humanitarian sector continually faces organizational and operational challenges to respond to the needs of populations affected by war, disaster, displacement, and health emergencies. With the goal of improving the effectiveness and efficiency of response efforts, humanitarian innovation initiatives seek to develop, test, and scale a variety of novel and adapted practices, products, and systems. The innovation process raises important ethical considerations, such as appropriately engaging crisis-affected populations in defining problems and identifying potential solutions, mitigating risks, ensuring accountability, sharing benefits (...) fairly, and managing expectations. This paper aims to contribute to knowledge and practice regarding humanitarian innovation ethics and presents two components related to a value-sensitive approach to humanitarian innovation. First is a mapping of how ethical concepts are mobilized in values statements that have been produced by a diverse set of organizations involved in humanitarian innovation. Analyzing these documents, we identified six primary values (do-no-harm, autonomy, justice, accountability, sustainability, and inclusivity) around which we grouped 12 secondary values and 10 associated concepts. Second are two proposed activities that teams engaged in humanitarian innovation can employ to foreground values as they develop and refine their project’s design, and to anticipate and plan for challenges in enacting these values across the phases of their project. A deliberate and tangible approach to engaging with values within humanitarian innovation design can help to ground humanitarian innovation in ethical commitments by increasing shared understanding amongst team members, promoting attentiveness to values across the stages of innovation, and fostering capacities to anticipate and respond to ethically challenging situations. (shrink)

The notion of validity fulfils a crucial role in legal theory. In the emerging Web 3.0, Semantic Web languages, legal ontologies, and normative multi-agent systems are designed to cover new regulatory needs. Conceptual models for complex regulatory systems shape the characteristic features of rules, norms, and principles in different ways. This article outlines one of such multilayered governance models, designed for the CAPER platform, and offers a definition of Semantic Web Regulatory Models . It distinguishes between normative-SWRM and institutional-SWRM. It (...) also compares existing principles in privacy by design, linked open data , legal information institutes , and online dispute resolution . The article concludes by proposing the notion of Relational Law to summarize the ethical dimension of SWRM. Ethics are the only regulatory way to constitute a global space, out of the jurisdictional public domain set by national, international, or transnational law, and opposed to the private one. (shrink)

This paper proposes a roadmap for how privacy leakages from outsourced managed security services using intrusion detection systems can be controlled. The paper first analyses the risk of leaking private or confidential information from signature-based intrusion detection systems. It then discusses how the situation can be improved by developing adequate privacy enforcement methods and privacy leakage metrics in order to control and reduce the leakage of private and confidential information over time. Such metrics should allow for quantifying how much information (...) that is leaking, where these information leakages are, as well as showing what these leakages mean. This includes adding enforcement mechanisms ensuring that operation on sensitive information is transparent and auditable. The data controller or external quality assurance organisations can then verify or certify that the security operation operates in a privacy friendly manner. The roadmap furthermore outlines how privacy-enhanced intrusion detection systems should be implemented by initially providing privacy-enhanced alarm handling and then gradually extending support for privacy enhancing operation to other areas like digital forensics, exchange of threat information and big data analytics based attack detection. (shrink)