Objectives of the chair
Research on ethical artificial intelligence has primarily focused on value alignment —the challenge of ensuring that intelligent machines operate in accordance with human moral norms and values.
This approach assumes that morals are fixed: either prescriptive, as defined by experts, or descriptive, as uncovered by behavioral science.
While this perspective is valuable, it also limits exploration of AI’s potential to shape moral norms. New research is needed to account for the dynamic nature of moral norms and the role intelligent machines may play in reinforcing, challenging, or transforming them. The Center for Moral AI (MoAI) aims to study the impact of intelligent machines on human moral norms and values using a multidisciplinary approach that combines moral psychology, experimental economics, and computer science.
MoAI projects begin by identifying the moral values likely to be affected by a given technology and why, drawing on insights from moral psychology. They then use experimental economics to design incentive-compatible protocols that measure the technology’s impact on those values.
Finally, computer science is used to develop a simplified or prototype version of the technology to be used in experiments—often despite the fact that the technology does not yet exist. This proactive approach allows us to prepare society for upcoming technological shifts and guide AI development to avoid harmful outcomes.
Research objectives
- Uptake of sincerity detection AI
- Whistleblowing of unethical delegation to AI
- AI Moral profiling and social credit systems
- Cooperation with AI
- Authenticity of AI-mediated communication
Main scientific goals
- Prediction of emerging moral norms for interactiong with AI
- Estimation of moral equilibrium of machine delegation
- Identification of paths to acceptable and unacceptable social credit systems
- Estimation of demand for real-time AI filters in online communications