HRI Values
Values are what shapes our thoughts and drives our actions. Over the years, philosophers, opinionated thinkers, and scientists have elaborated countless theories and taxonomies of human values; among these are Schwartz’s Theory of Basic Human Values and the Moral Foundations Theory. However, if we consider values to be an expression of preferences about the world around us, there is no need to limit ourselves to a specific theory. Furthermore, we can include other kinds of preferences, such as those related to the interaction with AI systems and robots. This line of research explores which values play a role in human-robot interactions and the consequences of the latest technological advancements on our lives.
PUBLICATION
Concerns and Values in HRI
I conducted a scoping review of value discussions in HRI over the past two decades, identifying 26 recurring concerns, such as the impact of robots on human interactions, attachment to robotic companions, and algorithmic bias, all categorised under values like agency, connectedness, privacy, autonomy, equity, and dignity. Visit the HRI Value Compass to explore these values and concerns in detail, and to find resources for designing robots that align with human values.
PUBLICATION
Conversational Jailbreaking Strategies
People will inevitably try to abuse systems built with LLMs. How do they try to do so when LLMs are powering social robots? With a Wizard of Oz approach we collect attempts to break an LLM-powered Misty robot and we categorise the techniques used, which include threats, role-play and the use of emotionally-charged language.