Knudsen and Munch awarded 3.2 million DKK for research on collective moral responsibility and AI
Nicolai Knudsen and Lauritz Munch has received 3.2 million DKK from the Independent Research Fund Denmark for a new research project on collective moral responsibility and artificial intelligence.
The project, titled Hybrid Agents and Moral Responsibility (HARM), addresses critical "responsibility gaps" that arise when AI systems cause harm, and it is unclear who is morally accountable. For example, if an advanced AI alters its own algorithm and overlooks a life-threatening diagnosis during a medical check-up, who is to blame? The doctor? The developer? Or no one at all?
These gaps raise both legal and ethical challenges. Current legal frameworks often hold individuals accountable in ways that may be morally unjust. To address this, HARM explores how AI systems can be understood as members of larger organizations or companies, forming what are called "hybrid agents." The project hypothesizes that hybrid agents can be morally responsible for the actions of their AI members, much like companies are responsible for the behavior of their employees.
By integrating insights from social ontology and AI ethics, HARM aims to provide a new framework for understanding moral responsibility in the age of AI. The findings will offer guidance on the responsible use of AI technologies and highlight potential needs for regulatory reform.
The HARM projet will run from March 2025 to October 2028.