From Ethics Oversight to Integration: Reshaping the Role of IRBs for AI Research [VIDEO]

Cansu Canca, Director of Responsible AI Practice, presented her talk "From Ethics Oversight to Integration: Reshaping the Role of IRBs for AI Research" in a Expeditions in Experiential AI Seminar on Wednesday, Feb. 14, 2024 at Northeastern University and online.

Abstract

With the developments in artificial intelligence (AI), the need for an ethical framework for AI research becomes pressing. In recognition of this need, a growing opinion in academia, research industry, and policy discussions has been favoring a system that is modeled after the research ethics framework in bioethics. This framework is one of ethics oversight and compliance, where institutional review boards (IRBs) act as authorities evaluating the ethical aspects of research protocols guided by a set of principles. Cansu argue that this IRB system’s existing challenges only get worse as it expands to evaluate AI systems. IRB system is ill-equipped to handle AI-specific moral questions and keep with the pace of AI research. Looking at the type of ethical issues that AI technology poses and on whom the risk of harm fall, the needed ethics framework goes well beyond research and encompasses the whole innovation lifecycle from research, development, and design to monitoring and updating processes. And the organizational structure to implement such a framework goes beyond the expertise, capacity, and scope of IRBs. Instead of acting as ethics authorities evaluating AI systems, she propose a function for IRBs as requiring an ethics integration and risk mitigation plan for research protocols and overseeing the robustness and the competence of the proposed plan and mechanism for the AI innovation lifecycle.

Biography

Cansu Canca is the Director of Responsible AI Practice at the Institute for Experiential AI and Research Associate Professor in Philosophy at Northeastern University. She is also the Founder and Director of AI Ethics Lab, one of the first initiatives focusing exclusively on advising practitioners and conducting multidisciplinary research on AI ethics. She serves as AI ethics and governance expert consultant to the UN and the INTERPOL, a member of the World Economic Forum’s AI Governance Alliance working group, a founding editor for AI & Ethics journal, and an ethics expert for EU and NIH funded research projects. She has a Ph.D. in philosophy specializing in applied ethics. She primarily works on ethics of technology, having previously worked on ethics and health.

‍