As large language models (LLMs) are embedded into mental health technologies, they are often framed either as tools assisting therapists or autonomous therapeutic systems. Such perspectives overlook their potential to mediate relational complexities in therapy, particularly for systemically marginalized clients. Drawing on in-depth interviews with 12 therapists and 12 marginalized clients in China, including LGBTQ+ individuals or those from other marginalized backgrounds, we identify enduring relational challenges: difficulties building trust amid institutional barriers, the burden clients carry in educating therapists about marginalized identities, and challenges sustaining authentic self-disclosure across therapy and daily life. We argue that addressing these challenges requires AI systems capable of actively mediating underlying knowledge gaps, power asymmetries, and contextual disconnects. To this end, we propose the Dynamic Boundary Mediation Framework, which reconceptualizes LLM-enhanced systems as adaptive boundary objects that shift mediating roles across therapeutic stages. The framework delineates three forms of mediation: Epistemic (reducing knowledge asymmetries), Relational (rebalancing power dynamics), and Contextual (bridging therapy-life discontinuities). This framework offers a pathway toward designing relationally accountable AI systems that center the lived realities of marginalized users and more effectively support therapeutic relationships.
Significant advancements have occurred in the application of Large Language Models (LLMs) for social simulations. Despite this, their abilities to perform teaming in task-oriented social events are underexplored. Such capabilities are crucial if LLMs are to effectively mimic human-like social behaviors and form efficient teams to solve tasks. To bridge this gap, we introduce MetaAgents, a social simulation framework populated with LLM-based agents. MetaAgents facilitates agent engagement in conversations and a series of decision making within social contexts, serving as an appropriate platform for investigating interactions and interpersonal decision-making of agents. In particular, we construct a job fair environment as a case study to scrutinize the team assembly and skill-matching behaviors of LLM-based agents. We take advantage of both quantitative metrics evaluation and qualitative text analysis to assess their teaming abilities at the job fair. Our evaluation demonstrates that LLM-based agents perform competently in making rational decisions to develop efficient teams. However, we also identify limitations that hinder their effectiveness in more complex team assembly tasks. Our work provides valuable insights into the role and evolution of LLMs in task-oriented social simulations.
Technology-mediated communication tools are widely used to maintain connections between older adults and their remote family members. However, challenges often arise due to differences in their life rhythms and communication preferences, such as time zones, daily routines, or preferred platforms. To address these challenges, we propose using Ditto, a mimetic embodied agent, in video-call-like interactions between an older adult and a remote family member. When direct interaction is difficult, FamilyDitto can represent either party, providing a strong social presence and a personalized experience to the other person. To explore the potential of Ditto in supporting intergenerational communication, we conducted seven co-design workshops (n=27) with older adults and younger family members. Our thematic analysis reveals Ditto's potential roles as both a temporal bridge and an emotional proxy, identifies the tension between faithful and idealized representations, and emphasizes the importance of personalization to support unique family dynamics. We distill our findings into implications for designing mediated semi-synchronous communication, using idealized representation responsibly, and addressing the asymmetric motivations and comfort levels with Ditto across roles and ages. This study provides a foundation for mimetic AI technologies that enhance, rather than replace, human connection in remote intergenerational relationships.
Several social networking sites offer automated machine translation of posts, but authors usually have no access to view or modify these translations. This may increase authors' concerns about whether the translations convey their intended meaning. To address this issue, we test a theory-driven model about human-in-the-loop translation using an online between-subjects experiment (N = 216). In our study, participants write fictitious social media status updates in a language other than English, which are then translated with one of three levels of automation: machine-provided translation with no author modifiability (MPT), where authors can view the machine translation but cannot modify it; author-provided translation (APT), where authors manually write the translation with no machine assistance; and machine-provided translation with author modifiability (MPT-AM), where authors can edit the machine translation of their post. We collect objective and subjective measures of users' experience using the assigned translation feature. The results from a structural equation modeling (SEM) analysis demonstrate that, compared to the MPT and APT conditions, participants in the MPT-AM condition reported higher measures of perceived control, ease of use, and perceived comfort, which in turn predicted higher perceived system effectiveness and ultimately increased intention to use the MPT-AM translation feature. Follow-up interviews (N = 15) found that participants appreciate the ability to edit and/or remove translations from their posts, even if they did not always do so.
The use of drones in public safety operations offers police departments potential benefits such as safety and efficiency. However, rarely considered are the perspectives of the public who are affected by the usage of drone in police operations. Ensuring and enhancing the sense of public safety is a primary goal of the police department. Therefore, public safety agencies and drone designers should consider the perception of bystanders at police operations and the broader society to ensure the fulfilment their primary purpose and maximizing drones' potential. This paper examines the factors that affect the public perception of safety during a field trial involving 80 test participants. Analysis of the group interviews reveal that unknowns are the main trigger of the sense of insecurity, including unknowns of drones' identity, purpose and usefulness. We found the public feels safer when the drones play an active role in delivering knowledge rather than passive role of data collection. Our findings highlight key design considerations for future drone-human interactions, providing insights into specific interaction features aimed at enhancing public perceptions of safety. We discuss the social and design implications to inform the public in solving the sense of insecurity mainly induced by a lack of knowledge.
A new class of technology professionals is shaping policy, informing legal arguments, and bolstering advocacy efforts from inside nonprofit and civil society organizations. This career path might be claimed by a number of different new sociotechnical domains: public interest technology (PIT), civic technology, data for good, technology for social justice, and others. Yet it is still unclear exactly what professional roles are emerging, what sorts of people are filling them, and what such individuals' work looks like and achieves. This work presents an interview study that seeks to characterize a specific sub-population of technological practitioners who are contributing materially to mission-driven projects from within the civil society or nonprofit sector: advocacy technologists. I present four patterns of praxis (i.e., professional practices and paradigms) common to advocacy technologists: their disposition as critics who interrogate technological paradigms and who introspect on their own ethical footprint, and their professional position translating between technical and non-technical worlds and trailblazing into new career paths. These four patterns demonstrate that advocacy technologists are choosing to occupy a precarious new niche within advocacy work ecosystems that has great potential to impact policy and design outcomes. Indeed, these practitioners enlist computational strategies to advance advocacy goals, situate deep sociotechnical expertise within policymaking contexts, and further civil society as an active site of tech design in its own right. This study contributes to the growing body of literature in human-computer interaction (HCI) and computer-supported cooperative work (CSCW) that explores computing technologies' role in processes and places of sociopolitical change. Ultimately, this work proposes that mission-driven civil society organizations and their technologists are not only underexplored sites for HCI and CSCW research, but also potentially rich collaborators for sociotechnical researchers who seek to deepen their impact on policy and social change.
Abusers routinely use technology to spy on and harass their targets. This harmful behavior is known as technology-facilitated abuse, or tech abuse. Survivors of tech abuse may turn to the legal system for safety and security, and to do so, they need evidence of tech abuse. However, prior work indicates challenges to collecting evidence of tech abuse and using it in legal proceedings. Thus, in this work, we study legal evidence used by survivors of tech abuse in Wisconsin, USA. We report on qualitative interviews and focus groups with 19 legal support providers who work with survivors seeking protective orders, divorces, and criminal charges. Our findings surface current practices that survivors and legal support providers use to prepare and present evidence of tech abuse in Wisconsin and the challenges they face. For example, survivors struggle to collect evidence of covert monitoring and surveillance. When they can collect evidence, it is often difficult to connect that evidence to the abuser due to the anonymous nature of many forms of tech abuse. In court, evidence of tech abuse is frequently challenged and vulnerable to objections and counter-evidence. And at the end of a proceeding, it's not uncommon for a judge to determine that the tech abuse does not meet the statutes. Informed by these results, we encourage CSCW and HCI researchers to work towards designing and deploying sociotechnical solutions that support survivors' use of evidence, in careful collaboration with advocates, legal experts, and survivors.
Youth experiencing housing insecurity or living in foster care have complex social, emotional, and material needs. While policies and structures exist to support these youth, intra-and interorganizational challenges pose barriers to effective collaboration. In contrast to prior work that focuses on the collaborative practices of individual stakeholders, this study draws on the insights of practitioners in different roles and organizations within the same county. Through 12 in-depth interviews with practitioners working across the system, we identify the key activities and infrastructural work performed by practitioners to support housing insecure and foster youth in light of social, technical, and institutional constraints. Based on our findings, we shed light on stigma management at an organizational level and the role of human infrastructure in responding to time-sensitive needs in highly institutionalized contexts.
Hospitals and care homes devote significant resources to placing post-acute patients from hospitals into long-term care. This paper describes a two-phase experiment over SMS, conducted with a hospital in Hawai'i, in which care homes express preferences, indicate availability to accept patients, and express interest in patients. In the first phase, the treatment asks care homes to reconsider their stated preferences to better support matching. The second phase measures whether resulting changes in preferences increased how often homes express interest in patients that match newly stated preferences. First, to motivate and inform experiment design, we explore factors contributing to extended hospital stays for patients, uncovering how care homes' preferences play a major role in what patients they consider accepting. Second, we conduct a 16-week randomized controlled trial with 960 homes, where we experimentally probed, via SMS messages, homes' willingness to change their preferences to improve potential patient match recommendations. We show that inducing homes to reflect on their preferences increased the number of homes who changed their preferences by over 50%: 9.8% of homes who received our treatment changed their preference compared to 6.0% of homes in the control group (p-value = 0.0421). Third, followup interviews with 22 home operators highlight how preference malleability is shaped by a combination of design constraints and on-the-ground realities, such as load-balancing existing patient rosters. Finally, we discuss implications for real-world systems like ours that must balance constrained communication with situational complexity towards improving outcomes.
The importance of human-centred explainable artificial intelligence (XAI) has been widely recognised, leading to a growing focus on users and practitioners during the explanation design and deployment processes. Previous studies have identified that users with different domain expertise may have diverse needs for explainability. However, current XAI research often conflates a user's practical experience with domain expertise, ignoring the distinctions between the two; generally, experience relates to acquiring skill and insight through active participation or observation, while domain expertise denotes a high level of (often highly local and/or specific) knowledge. This paper investigates the impact of users' practical experience and domain expertise on how AI recommendations are considered in a high-risk decision-making context, using the example of ball bearing fault diagnosis in the manufacturing sector. As an interdisciplinary team of human-computer interaction (HCI) researchers and mechanical engineers, we co-design an XAI-based simulated ball bearing fault diagnostic task. We conduct task-led interviews with several professionals, structured around three distinct decision processes, and use an innovative sketch-based exercise to gather data to demonstrate how their decision making behaviours change under ML recommendations and AI explanations. Our results show that highly experienced and knowledgeable practitioners understand but rely less on the explanations, while those with high experience but low expertise are more easily misled. Practitioners with high expertise but low experience trust XAI but struggle to use the explanations effectively. Based on these observations, we reflect on our methods and argue for considering both domain expertise and practical experience when designing and deploying AI explanations.
Surveys are a powerful tool for collecting data and eliciting insights on social phenomena, and are critical in product design, marketing, scientific research, and other domains. However, traditional open-ended and closed-ended question formats limit researchers' ability to capture data that combines both the richness of qualitative insights and the analytical rigor of quantitative data. Closed-ended questions facilitate structured data collection that is amenable to statistical analysis but limit respondents' answers. In contrast, open-ended questions allow for nuanced responses incorporating new perspectives but require significant effort to interpret due to their unstructured nature. Moreover, traditional survey tools lack mechanisms to prompt respondents for deeper reflections or to facilitate engagement with others' perspectives, limiting the potential for richer insights. To address these problems, we propose Dynamic Surveys, a survey platform that uses Large Language Models (LLMs) to dynamically cluster qualitative responses in real time and to elicit quantitative ratings and rankings on those clusters and qualitative reflections on how their views compare to broader respondent trends, especially helpful in early-stage or exploratory research settings. This process generates a report showing survey creators and respondents the clustered responses as well as each cluster's rank, rating distribution, and follow-up reflections. To evaluate Dynamic Surveys, we conducted two field studies with 93 participants over a 2-month period. In the first study, 52 students provided input for a career workshop, while in the second, 41 students gave feedback on gaps in their academic curriculum. Of these, 44 respondents filled out a survey on their experience using Dynamic Surveys. We also shared the generated report with 4 individuals who were interested in the insights for their work, and interviewed them to understand their perspectives on the results and any contextual risks they saw in the platform design. Our findings suggest that Dynamic Surveys not only provide richer and deeper insights into responses compared with traditional survey tools, but also increase engagement and foster a sense of community. We discuss broader implications for the design of survey platforms that blend qualitative depth with quantitative structure, facilitating richer insights and offering more collaborative interactions.
Despite the absence of consent being a defining quality of computer-mediated sexual harm, there is an absence of consent models that explicitly prescribe how consent to sexual activity should be asked for, given, and denied when mediated by technology. HCI literature has advocated for the adoption of affirmative consent ("yes means yes"); however, this model was created in 1991 without consideration for computers and has been historically underutilized. Through a speculative study of VR dating with 16 women and LGBTQIA+ stakeholders, we contribute archetypes of four new computer-mediated consent models for sexual activity. These include 1) visual consent through AR/VR rather than verbal dialogue, 2) agent-mediated consent where AI agents communicate consent on behalf of sexual partners, 3) a two-layer consent process called consent-to-stimulus, and 4) environmental consent where virtual environments scaffold behaviors that can(not) be consented to. We conclude by reflecting on which models could potentially supplant affirmative consent to better mitigate computer-mediated sexual violence and harassment. Content warning: This paper discusses forms of sexual violence including rape.
Through interviews with 16 social justice activists, we explore their challenges of adapting to Instagram, particularly in light of the platform's evolving algorithm. Our findings reveal that the frequent changes in these algorithms significantly impact their ability to engage effectively-and disproportionately impact visibility, especially for those with fewer resources and less algorithmic expertise. Our contributions encompass discussions on activists' challenges in adapting to platform changes, and the strategic shifts towards gaining broader visibility. We also address the expectations of being a "good digital activist" amidst algorithmic mediation on Instagram, emphasizing participants' need for navigating platform mediated complexities and maintaining authenticity. Finally, we suggest design implications, advocating features-for both existing platforms and alternative systems exclusively for activism that reduce activists' concerns about quantitative metrics, promote selective privacy, tie amplification to thoughtful engagement, and foster community building through contextual moderation and communication.
Marginalized groups often face situations in which their knowledge and experiences are dismissed due to prejudice or bias-a phenomenon identified and theorized as epistemic injustice in feminist philosophy. These circumstances frequently compel individuals to produce additional evidence to support their claims, ranging from paper documentation to data generated by technologies such as location logs. This paper examines the case of Heat Seek, an internet-connected temperature sensor designed to provide tenants in New York City with "objective and reliable data" when filing heating complaints and appearing in housing court. We present findings from a qualitative study, supplemented by document review and artifact analysis, to illuminate the tool's functions and uses. Drawing on this case, we introduce a class of civic technologies-credibility boosters. We find that these technologies aim to overcome credibility deficits by: (1) backing individual and collective claims with objective data, (2) materializing intangible experiences as tangible evidence with aesthetic reliability, and (3) shifting epistemic authority to perceived neutral third parties. We conclude by demonstrating the institutional and social impacts of such technologies and call for greater attention to epistemic injustices within CSCW research, advocating for the design of institutional, legal, and social systems that confront biased systems and empower marginalized communities.
In August 2020, Alexander Lukashenko's re-election, amid widespread allegations of electoral fraud, marked the continuation of his uninterrupted presidency since Belarus's independence and triggered an unprecedented wave of mass protests in the country's history. In response, Lukashenko's adaptive authoritarian regime unleashed brutal repression and systemic human rights violations. In this context, the diasporic social movement community, leveraging information and communication technologies (ICTs), emerged as critical actors supporting the anti-regime social movement in their origin-homeland. Based on semi-structured interviews with 13 members of the North American Belarusian diasporic social movement community, this paper explores the role of ICTs in facilitating their political actions during the 2020 protests, as well as the factors that facilitated or hindered their participation and use of ICTs. Our study highlights that ICTs facilitated diaspora geopolitics from below by enabling "social movement community," where otherwise disparate diasporic satellite publics converged around the common political goal of overthrowing the Lukashenko regime. However, the diasporic social movement community's use of ICTs was also fraught with ethical and moral complexities, navigating the "proximity dilemma" of remote participation and influencing a cause from a distance, while benefiting from socio-spatial privileges in their host country. Furthermore, the diaspora's ICT usage is shaped by the political regime, fear of transnational repression, and the geopolitical positions of both the origin-homeland and host country, as a consequence of adaptive authoritarianism in the Belarusian case. We discuss how CSCW can support decentralised, geographically dispersed diasporic organising with respect to social movements under varying authoritarian constraints.
Our work explores how audiences perceive and engage with remote music events. The shift to digital platforms has transformed the live music experience, leading to the rise of remote music performances. Following the COVID-19 pandemic, platforms such as Twitch have been repurposed and surged in popularity as meaningful outlets for live music dissemination. Concurrently, novel platforms specifically crafted for music events have begun to emerge. Understanding the preferences and challenges of remote music audiences is crucial because it guides the development of platforms that effectively answer varied user needs, which could enhance overall engagement and shape the future of music consumption. For this, we employed a qualitative multi-method approach, including a survey, interviews, and event observations, to start mapping the design space of remote music performances and capture a wide spectrum of audience perspectives. Our findings reveal audience engagement patterns in remote music events across different digital platforms, highlighting unmet needs, desires as well as opportunities for future HCI work in this area.
The creative internet is a network of online platforms, creative people, and communities where artists learn and grow together while also showcasing their artwork to others. While this space of people, technology, and routine presentation of art once held such promise for creatives, for many, that place is something that now only exists in memory. This paper presents an interview study with 22 visual artists who share their art in online social spaces. Taking a historical approach, we explore their journeys across the creative internet, the promise it held for them as artists, and how the creative internet became enshittified as online platforms got bigger. Through the lens of enshittification, we explore how changes in platform policy, design, and algorithmic mediation have shifted the creative practices of artists, and how, as artists contend with enshittified platform spaces, they are resilient and forward joy in their art. We discuss the role that nostalgia plays in the concept of enshittification, contributing insight into why people stay on increasingly hostile platforms, and suggest opportunities for platforms and artists alike to forward joy and resilience in the face of a now shittier internet.
The proliferation of (semi-)autonomous and learning interactive systems (i.e., artificial intelligence-driven, AI) will significantly change work practices. While AI promises an immense increase in performance, it also has novel experiential, social, cultural, and ethical implications. Unlike conventional systems, AI's potential agency and decision-making capabilities will turn AI-driven systems into counterparts and, at best, collaborators. We used a speculative, participatory approach to collect narratives about current and future work practices with AI, their conditions, and consequences. This informed "WorkAI", a card deck and a canvas to support the design of AI-driven work. The twenty cards encourage participants to apply important aspects such as reliability/trust or meaning to collaboration with AI. The toolkit was developed in two iterations. It provided structure to the discussion and stimulated questions that facilitated the development of future narratives about working with AI.
YouTube has become an important part of the educational ecosystem, with millions of viewers seeking informative videos and help with coursework. Educational YouTubers create this content, often balancing pedagogical rigor and entertainment value. However, creators need not only to promote their content to find viewers, but also to monetize. In this study, we explore the tensions educational YouTubers face when making monetized educational content. We conduct a qualitative interview study with 12 popular educational YouTubers about their monetization strategies, perceptions of YouTube's algorithmic promotion of their content, and conception of their audience. We find that educational YouTubers are largely driven by a desire to share free and high-quality educational content, and that common monetization strategies like sponsorships and clickbait sometimes interfere with this mission. We describe the careful strategies our participants use to maintain educational integrity while making a living on an algorithmically-driven platform. We then use these findings to draw parallels between YouTubers' challenges with monetizing educational content and the history of educational public broadcast in the United States, which has followed a similar trajectory. In closing, we offer several recommendations for supporting educational YouTubers in creating the high-quality, publicly accessible educational content that is appreciated by a worldwide audience.
Remote collaborative design has become increasingly popular, but current design tools often overlook the importance of contextual communication during synchronized design activities, which is critical for understanding the rationale and decisions behind design choices. In this paper, we introduce DesignMemo, a proof-of-concept system that integrates the verbal context of remote discussions into visual design history tracking. The system automatically labels the visual elements with an annotation, which is linked to a certain transcript of the meeting, so that the user can easily recall the context of the visual design by clicking the element. The system also integrates an LLM agent for annotation-oriented summarization based on global context tracking, so users can quickly follow the rationale of the design without reading the lengthy transcript. Our user study with 24 participants suggests that the ability to track communication context makes the iterative design process smoother and more efficient.
