SODAS Data Discussion #1 (Spring 2024)
Copenhagen Center for Social Data Science (SODAS) aspirers to be a resource for all students and researchers at the Faculty of Social Sciences. We therefore invite researchers across the faculty to present ongoing research projects, project applications or just a loose idea that relates to the subject of social data science.
Two researchers will present their work. The rules are simple: Short research presentations of ten minutes are followed by twenty minutes of debate. No papers will be circulated beforehand, and the presentations cannot be longer than five slides.
Discussion 1
Presenter: Rosa Lavelle-Hill.
Title: Understanding Heterogeneity in AI Performance.
Abstract:
Algorithms are increasingly part of our daily lives, with their predictions mattering more and more [1]. On the one hand, algorithms can be used for social good, e.g., by providing early warning systems for diseases [2]. On the other hand, algorithms can be used to prompt people to spend money for goods that they initially might neither want, need, nor have money for, e.g., by using personalized persuasion [3]. Therefore, depending on the context, being highly (un)predictable as a person can be considered either advantageous or disadvantageous. Predictability has thus far largely been studied from statistical-computational perspectives (e.g., complex systems theory [5], information theory [4], learning theory [6]). In this data discussion, I argue that when human behavior is the outcome, there is also a need to study predictability from a psychological perspective. For instance, individuals with certain personality characteristics might show highly sporadic (and thus unpredictable) or, vice versa, compulsively repetitive (and thus predictable) behavior. To facilitate ethical, fair, and unbiased prediction outcomes for all people, it is essential to understand which characteristics relate to predictability and the social consequences of this. I will present future project plans (supported by the Carlsberg Foundation and the University of Basel) on understanding heterogeneity in AI performance from a psychological and ethical perspective.
Discussion 2
Presenter: Kristin Eggeling.
Title: Ethnographic Insights Into the Digitalisation of Diplomacy.
Abstract:
Screens are everywhere in the modern world, including in international politics. In this article, I consider how diplomacy, a profession traditionally premised on face-to-face interactions, is adjusting to life with and on the screen. For this, I present ethnographic material from five years (2018-2023) of research in and with the diplomatic scene in Brussels. I approach this material through the work of Sherry Turkle and apply her ideas about ‘evocative objects’ to the presence of screens in diplomacy. In contrast to existing scholarship that asks in a more abstract ways about how digital technologies change diplomacy, I ask more concisely about what kind of associations screens (e.g., smartphones) evoke among diplomatic professionals. Based on this, I draw out how screens confirm or confront established professional procedures, and theorise the screen as an expressive tool that both catches and shapes the rules and frictions of the diplomatic field. Overall, the article contributes detailed empirical insights into an otherwise largely closed-off world, and nurtures a new theoretical language about one of the key objects that drives the digitalisation of international politics.