Artificial intelligence

AI research is growing rapidly raising various practical ethical issues related to safety, risks, and other effects that are being widely discussed by researchers worldwide. We aim at further developing the theoretical reflection about AI that is needed to articulate feasible conceptual and ethical tools for assessing AI-related research within the Human Brain Project, as well as its potential applications. With the goal of ensuring that this work is ethically sustainable and will benefit society.

Practically oriented reflection usually includes listing a few fundamental principles that are ´classical´ in the bioethics literature (e.g., human dignity, respect for autonomy, non-maleficence, beneficence) and adding other relevant considerations (such as respect for  human rights, promotion of human well-being and flourishing, transparency, accountability, and effectiveness, among others). Together, they are assumed to provide an adequate basis for achieving an ethically sound development and application of AI. However, it is important to note that there is a tendency in the relevant literature and documents to rely on the view that ethical reflection on AI is a subfield of applied ethics, either implicitly or explicitly.  While subsuming AI ethics under the umbrella of applied ethics covers important aspects of the ethical reflection on AI, the normative discussion of important applied issues (e.g., privacy, data protection, the impact of AI on job market, etc.) is insufficient without the examination of key concepts and categories.

Among the topics we are discussing, we are reflecting specifically on a number of issues, e.g. definition of AI ethics as more than applied ethics, reflection on how much our reflection on AI is biased by an anthropomorphic attitude, research on criteria for ascribing consciousness to AI.

In the context of our work within the European Human Brain Project, our research focuses on:

• Reflecting on the definition and the methodology of AI ethics
• Developing criteria for an ethical analysis of AI
• Developing criteria for ascribing consciousness to AI
• Exploring the ethical implications of our biases (e.g., anthropomorphism) about AI

Contact

• Michele Farisco, PhD, Post-Doc Researcher

Collaborators

• Arleen Salles, PhD, Senior Researcher
• Kathinka Evers, PhD