A person deciding whether to let an AI triage their hospital care is making a moral judgment about a machine—and this project will find out what drives that trust. This matters because AI systems are already being asked to make or influence decisions with ethical weight, from allocating ventilators to prioritising social care visits. Yet we know surprisingly little about when and why humans actually trust these "moral machines." The gap is not just technical—it is psychological and philosophical. TRUST-AI will investigate four core questions: what features of an AI make it seem trustworthy, what personal differences affect willingness to trust, what situations lead people to hand over moral decisions, and how these insights should shape AI design. If successful, the research could change how AI systems are built and regulated. Rather than assuming trust follows from accuracy or transparency, developers would have evidence on what actually earns human confidence in morally charged contexts. The project will run online experiments in at least three languages, conduct cross-cultural studies in nine countries, and test roughly 29,000 participants across 20 studies. The findings will inform how autonomous systems are deployed in healthcare, criminal justice, and other settings where moral stakes are high.
View original technical description
Artificial intelligence (AI) is increasingly used to perform tasks with a moral dimension, such as prioritising scarce medical resources. Debates rage about the ethical issues of AI, how we should programme ethical AI, and which ethical values we need to prioritize. But machine morality is as much about human moral psychology as it is about the philosophical and practical issues of building artificial agents. To reap the benefits of AI, stakeholders need to be willing to use, adopt, and rely on these systems: they must trust in the AI agents. TRUST-AI draws on psychology and philosophy to explore how and when humans trust AI agents that act as 'moral machines'. Drawing from classic models of trust and recent theoretical work from moral psychology on the complexity of trust in the moral domain, this five-year project explores 1) The characteristics of AI agents that predict trust; 2) the individual differences that make us more or less likely to trust AI agents; 3) the situations where we are more likely to 'outsource' moral decisions to AI agents; and 4) how these findings should be used to design AI agents that warrant our trust. My approach is methodologically pluralistic and includes qualitative analysis and natural language processing, behavioural economic games, and in-person behavioural experiments. I develop a customised data collection platform that will run online experiments globally in at least 3 different languages, as well as cross-cultural studies in 9 different countries and a range of experiments with an estimated 29,000 participants across four work programmes and 20 studies. These findings will help us understand the how, when, and why people trust AI agents with important theoretical and methodological implications for research on the antecedents and consequences of trust in moral machines.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know