Search for: All records

Access to mortality data is critical for researchers for various reasons, including the development of life-saving medical interventions. However, high-quality mortality data is typically unavailable for researchers to access in a timely manner due to bottlenecks and inefficiencies in the coordination of managing mortality data across a vast information ecosystem. To identify opportunities to improve access to mortality data, we conducted a qualitative interview study with 20 experts with direct knowledge and experience with mortality data systems. From these interviews, we mapped the mortality data ecosystem and found that policy constraints are the main underlying cause of bottlenecks and inefficiencies. We argue that policy intervention is required to optimize the coordination of mortality data between databases and between organizations. To support researcher access to mortality data, we present guiding principles for designers seeking to improve the mortality data ecosystem. These principles contribute to CSCW scholarship focused on the challenges of coordination across large information ecosystems and the tradeoffs between technology and policy when designing such systems. 

The rapid development and deployment of generative AI technologies creates a design challenge of how to proactively understand the implications of productizing and deploying these new technologies, especially with regard to negative design implications. This is especially concerning in CSCW applications, where AI agents can introduce misunderstandings or even misdirections with the people interacting with the agent. In this panel, researchers from academia and industry will reflect on their experiences with ideas, methods, and processes to enable designers to proactively shape the responsible design of genAI in collaborative applications. The panelists represent a range of different approaches, including speculative fiction, design activities, design toolkits, and process guides. We hope that the panel encourages a discussion in the CSCW community around techniques we can put into practice today to enable the responsible design of genAI. 

Fairness metrics have become a useful tool to measure how fair or unfair a machine learning system may be for its stakeholders. In the context of recommender systems, previous research has explored how various stakeholders experience algorithmic fairness or unfairness, but it is also important to capture these experiences in the design of fairness metrics. Therefore, we conducted four focus groups with providers (those whose items, content, or profiles are being recommended) of two different domains: content creators and dating app users. We explored how our participants experience unfairness on their associated platforms, and worked with them to co-design fairness goals, definitions, and metrics that might capture these experiences. This work represents an important step towards designing fairness metrics with the stakeholders who will be impacted by their operationalizations. We analyze the efficacy and challenges of enacting these metrics in practice and explore how future work might benefit from this methodology. 

Artificial intelligence (AI) has become an increasingly critical component of not only the computing workforce but also society. It is essential for a diverse group of young people to contribute to this field. However, even within computing, AI is not taught to all post-secondary students. Students often must self-select into AI courses, meaning their reasons for choosing AI may be based on preconceptions of the discipline that may or may not be accurate. We extend the work of a small-n interview study of primarily Asian/Asian American undergraduate students, many of whom expressed perceptions of AI that paralleled identified computing stereotypes. Many of these stereotypes have the potential to discourage undergraduate computing students to take classes or specialize in AI, particularly those from underrepresented groups. Here we present a larger scale validation of those findings in the form of survey data conducted at a large public research institution in the USA. The survey largely confirmed the findings of the interview study at a larger scale, and we also found that gender did not significantly influence the results. Finally, we discuss strategies for AI integration into non-AI computing courses based on those previously used in responsible computing contexts, the goal being to counter harmful preconceptions before students specialize into computing subareas. 

Cybersecurity expertise continues to be relevant as a means to confront threats and maintain vital infrastructure in our increasingly digitized world. Public and private initiatives have prioritized building a robust and qualified cybersecurity workforce, requiring student buy-in. However, cybersecurity education typically remains siloed even within computer and information technology (CIT) curriculum. This paper's goal is to support endeavors and strategies of outreach to encourage interest in cybersecurity. To this end, we conducted a survey of 126 CIT students to investigate student perceptions of cybersecurity and its major crosscutting concepts (CCs). The survey also investigates the prevalence of preconceptions of cybersecurity that may encourage or dissuade participation of people from groups underrepresented in computing. Regardless of prior learning, we found that students perceive cybersecurity as a relatively important topic in CIT. We found student perspectives on conceptual foundations of cybersecurity were significantly different (p < .05) than when simply asked about "cybersecurity," indicating many students don't have an accurate internal construct of the field. Several previously studied preconceptions of cybersecurity were reported by participants, with one misconception - that cybersecurity "requires advanced math skills" - significantly more prevalent in women than men (p < .05). Based on our findings, we recommend promoting cybersecurity among post-secondary students by incorporating elements of cybersecurity into non-cybersecurity CIT courses, informed by pedagogical strategies previously used for other topics in responsible computing. 

AI technologies are likely to impact an array of existing practices (and give rise to a host of novel ones) around end-of-life planning, remembrance, and legacy in ways that will have profound legal, economic, emotional, and religious ramifications. At this critical moment of technological change, there is an opportunity for the HCI community to shape the discourse on this important topic through value-sensitive and community-centered approaches. This workshop will bring together a broad group of academics and practitioners with varied perspectives including HCI, AI, and other relevant disciplines (e.g., law, economics, religious studies, etc.) to support community-building, agenda-setting, and prototyping activities among scholars and practitioners interested in the nascent topic of how advances in AI will change socio-technical practices around death, remembrance, and legacy.