Daniel Sullivan

Jack Manning, Daniel Sullivan, Dylan Thomas Doyle et al.

We examine how people experience two choices in the design of generative ghosts, AI systems that are trained on data of the dead: representation, where an AI speaks about a deceased person in the third person, and reincarnation, where the AI speaks as the deceased in the first person. Through a qualitative user study with 16 participants, we explore how each shaped authenticity, affect, and risk. Reincarnation was preferred for its immediacy, but participants shared fears of over-reliance. Representation was preferred for engaging with memory over conversational presence, though participants often ignored this distinction, engaging in dialogue despite third-person framing. Across both modes, participants privileged affective resonance over factual fidelity. We conclude by showing how factors such as tone, language, and conversational rhythm -- factors unique to the user's memory of the deceased -- shape interactions with generative ghosts, and argue that those interactions are always collaborative.

Daniel Sullivan, Carlos Caminha, Victor Dantas et al.

Listing people alphabetically on an electronic output device is a traditional technique, since alphabetical order is easily perceived by users and facilitates access to information. However, this apparently harmless technique, especially when the list is ordered by first name, needs to be used with caution by designers and programmers. We show, via empirical data analysis, that when an interface displays people's first name in alphabetical order in several pages/screens, each page/screen may have imbalances in respect to gender of its Top-k individuals.k represents the size of the list of names visualized first, which may be the number of names that fits in a screen page of a certain device.The research work was carried out with the analysis of actual datasets of names of five different countries. Each dataset has a person name and the frequency of adoption of the name in the country.Our analysis shows that, even though all countries have exhibit imbalance problems, the samples of individuals with Brazilian and Spanish first names are more prone to gender imbalance among their Top-k individuals. These results can be useful for designers and engineers to construct information systems that avoid gender bias induction.