The organization of behavior: A neuropsychological theory

After virtue, a study in moral theory

In the past five years, private companies, research institutions and public sector organizations have issued principles and guidelines for ethical artificial intelligence (AI). However, despite an apparent agreement that AI should be ‘ethical’, there is debate about both what constitutes ‘ethical AI’ and which ethical requirements, technical standards and best practices are needed for its realization. To investigate whether a global agreement on these questions is emerging, we mapped and analysed the current corpus of principles and guidelines on ethical AI. Our results reveal a global convergence emerging around five ethical principles (transparency, justice and fairness, non-maleficence, responsibility and privacy), with substantive divergence in relation to how these principles are interpreted, why they are deemed important, what issue, domain or actors they pertain to, and how they should be implemented. Our findings highlight the importance of integrating guideline-development efforts with substantive ethical analysis and adequate implementation strategies.

The global landscape of AI ethics guidelines

In the last five years, private companies, research institutions as well as public sector organisations have issued principles and guidelines for ethical AI, yet there is debate about both what constitutes "ethical AI" and which ethical requirements, technical standards and best practices are needed for its realization. To investigate whether a global agreement on these questions is emerging, we mapped and analyzed the current corpus of principles and guidelines on ethical AI. Our results reveal a global convergence emerging around five ethical principles (transparency, justice and fairness, non-maleficence, responsibility and privacy), with substantive divergence in relation to how these principles are interpreted; why they are deemed important; what issue, domain or actors they pertain to; and how they should be implemented. Our findings highlight the importance of integrating guideline-development efforts with substantive ethical analysis and adequate implementation strategies.

Artificial Intelligence: the global landscape of ethics guidelines.

In information societies, operations, decisions and choices previously left to humans are increasingly delegated to algorithms, which may advise, if not decide, about how data should be interpreted and what actions should be taken as a result. More and more often, algorithms mediate social processes, business transactions, governmental decisions, and how we perceive, understand, and interact among ourselves and with the environment. Gaps between the design and operation of algorithms and our understanding of their ethical implications can have severe consequences affecting individuals as well as groups and whole societies. This paper makes three contributions to clarify the ethical importance of algorithmic mediation. It provides a prescriptive map to organise the debate. It reviews the current discussion of ethical aspects of algorithms. And it assesses the available literature in order to identify areas requiring further work to develop the ethics of algorithms.

https://journals.sagepub.com/doi/pdf/10.1177/2053951716679679

The ethics of algorithms: Mapping the debate:

Intelligence agencies have identified artificial intelligence (AI) as a key technology for maintaining an edge over adversaries. As a result, efforts to develop, acquire, and employ AI capabilities for purposes of national security are growing. This article reviews the ethical challenges presented by the use of AI for augmented intelligence analysis. These challenges have been identified through a qualitative systematic review of the relevant literature. The article identifies five sets of ethical challenges relating to intrusion, explainability and accountability, bias, authoritarianism and political security, and collaboration and classification, and offers a series of recommendations targeted at intelligence agencies to address and mitigate these challenges. 

/pdf/the-ethics-of-artificial-intelligence-for-intelligence-ey0u05b9.pdf

The Ethics of Artificial Intelligence for Intelligence Analysis: a Review of the Key Challenges with Recommendations

The Doctrine of Double Effect & Lethal Autonomous Weapon Systems

Human Control of Artificial Intelligent Systems: A Critical Review of Key Challenges and Approaches

. The EU proposal for the Artificial Intelligence Act (AIA) defines four risk categories: unacceptable, high, limited, and minimal. However, as these categories statically depend on broad fields of application of AI systems (AIs), the risk magnitude may be wrongly estimated, and the AIA may not be enforced effectively. Our suggestion is to apply the four categories to the risk scenarios of each AIs, rather than solely to its field of application. We address this model flaw by integrating the AIA with the framework arising from the Intergovernmental Panel on Climate Change (IPCC) reports and related literature. This makes possible addressing AI risk considering the interaction between (a) risk determinants, (b) individual drivers of determinants, and (c) multiple risk types. Then we integrate the proposed model with a proportionality-based balance among values considered by the AIA’s risk analysis. The resulting semi-quantitative approach identifies a more efficient way to implement the AIA and addresses the regulatory issue of general-purpose AI (GPAI).

/pdf/how-to-evaluate-the-risks-of-artificial-intelligence-a-1dzgd4pf.pdf

How to Evaluate the Risks of Artificial Intelligence: A Proportionality-Based, Risk Model for the AI Act

Cybersecurity continues to be a key element of national security. Given that critical infrastructure is powered by digital systems, any disruption to the flow of vital services has the potential effect of causing harm to citizens, organizations, and government entities. The availability of service offerings may be subject to attacks by AI on various aspects of cyber-physical-social systems. And no system is foolproof. Inasmuch as AI can be a threat to cybersecurity, it can also be used to combat attempts at systems penetration through deep learning and other capabilities. However, focusing our attention on AI alone does not allow for a sufficiently accurate understanding of the complex ecosystem that contributes to the conflicting scenarios that are possible in the context of AI in cybersecurity. This Special Issue provides insights into the broader socio-technical ecosystem considerations that are required for the achievement of system robustness, allowing for a more intricate and balanced understanding of the role of AI in cybersecurity, at various levels relevant to individuals, organizations, and communities/ society. Fundamentally, the socio-technical approach calls for an emphasis on people, processes, and technology, and not solely on the shiny gadgetry that we call products. There are four main themes that form the emergent research agenda: 1) the cybersecurity ecosystem that demands an interdisciplinary approach; 2) the state of the art technology development processes and landscape that provide and facilitate cybersecurity mechanisms and competitive advantage through AI; 3) the interconnected and multifaceted nature of the social, technical, and environmental subsystems in the cybersecurity ecosystem; and 4) the emphasis on human requirements and values, inclusive of end-user capacity and awareness, organizational capability and skills development, and societal considerations inclusive of the public interest. 

/pdf/socio-technical-ecosystem-considerations-an-emergent-1nk4y1s8.pdf

Mariarosaria Taddeo

Papers

The Ethics of Artificial Intelligence for Intelligence Analysis: a Review of the Key Challenges with Recommendations

The Doctrine of Double Effect & Lethal Autonomous Weapon Systems

Human Control of Artificial Intelligent Systems: A Critical Review of Key Challenges and Approaches

How to Evaluate the Risks of Artificial Intelligence: A Proportionality-Based, Risk Model for the AI Act

Socio-Technical Ecosystem Considerations: An Emergent Research Agenda for AI in Cybersecurity