This study investigates how Iranian immigrant nonprofit groups experience exclusion within technology and AI-driven infrastructures. Based on 27 semi-structured interviews, it identifies how legitimacy barriers, capacity gaps, and ethical dilemmas intersect to create a cycle of infrastructural immobility that restricts these groups’ participation in the digital nonprofit ecosystem. The findings reveal that generative AI, while providing some opportunities, risks exacerbating these challenges by deepening marginalization and reinforcing inequalities in access, data visibility, AI fluency, and AI-mediated representation. Uncritical adoption of generative AI in the nonprofit domain undermines transparency and human connection in everyday organizational practice and induces bias in AI-generated content in the context of Iranian immigrant nonprofit work. To address these issues, the paper proposes interaction-level design strategies that promote community-driven inclusion, support context-aware capacity building, and leverage AI’s augmentative potential to strengthen transparency practices and human connection among Iranian immigrant nonprofits.

This paper examines how non-resident Bangladeshis mobilized during the 2024 quota-reform turned pro-democracy movement, leveraging social platforms and remittance flows to challenge state authority. Drawing on semi-structured interviews, we identify four phases of their collective action: technology-mediated shifts to active engagement, rapid transnational network building, strategic execution of remittance boycott, reframing economic dependence as political leverage, and adaptive responses to government surveillance and information blackouts. We extend postcolonial computing by introducing the idea of “diasporic superposition," which shows how diasporas can exercise political and economic influence from hybrid positionalities that both contest and complicate power asymmetries. We reframe diaspora engagement by highlighting how migrants participate in and reshape homeland politics, beyond narratives of integration in host countries. We advance the scholarship on financial technologies by foregrounding their relationship with moral economies of care, state surveillance, regulatory constraints, and uneven international economic power dynamics. Together, these contributions theorize how transnational activism and digital technologies intersect to mobilize political change in Global South contexts.

Sociotechnical systems, such as language technologies, frequently exhibit identity-based biases. These biases exacerbate the experiences of historically marginalized communities and remain understudied in low-resource contexts. While models and datasets specific to a language or with multilingual support are commonly recommended to address these biases, this paper empirically tests the effectiveness of such approaches for gender, religion, and nationality-based identities in Bengali, a widely spoken but low-resourced language. We conducted an algorithmic audit of sentiment analysis models built on mBERT and BanglaBERT, which were fine-tuned using all Bengali sentiment analysis (BSA) datasets from Google Dataset Search. Our analyses showed that BSA models exhibit biases across different identity categories despite having similar semantic content and structure. We also examined the inconsistencies and uncertainties arising from combining pre-trained models and datasets created by individuals from diverse demographic backgrounds. We connected these findings to the broader discussions on epistemic injustice, AI alignment, and methodological decisions in algorithmic audits.

Language operates as a mechanism of both marginalization and resistance, especially for minority communities navigating insensitive and harmful speech online. As content moderation increasingly depends on large language models (LLMs), concerns arise about whether these systems can recognize culturally insensitive speech–language that disregards or marginalizes the cultural and religious perspectives of historically underrepresented communities, often through implicit erasure, misrepresentation, or normative framing, rather than overt hostility. Focusing on Bangladesh’s Hindu and Chakma communities – the country’s largest religious and Indigenous ethnic minorities, respectively – this paper investigates the epistemic limits of LLM-based moderation systems and explores methods for incorporating minority perspectives. We co-created a culturally grounded corpus of insensitive speech with community members and integrated their narratives into moderation pipelines using retrieval augmented generation (RAG). Our tool, Mod-Guide, improves LLM sensitivity to minority viewpoints by leveraging contextual cues derived from lived experience. Through mixed-method evaluations involving both minority and majority participants, we demonstrate that RAG-enhanced moderation responses are more contextually accurate and perceived differently across ethnic lines. This work advances research in human-computer interaction, AI ethics, and social computing by foregrounding restorative justice and hermeneutical inclusion in the design of content moderation systems.

This paper examines how algorithmic accountability in Canada's visa system is articulated institutionally and experienced by applicants across borders. We analyzed Immigration, Refugees and Citizenship Canada (IRCC)'s Algorithmic Impact Assessment (AIA) for the temporary resident visa (TRV) triage system using the algorithmic decision-making adapted for the public sector (ADMAPS) framework and analyzed Reddit discussions among applicants using a mixed-methods approach. We show that while institutional artifacts emphasize transparency, procedural safeguards, and bounded impacts, applicants engage in collective sensemaking to interpret opaque decisions, often relying on peer knowledge amid uncertainty. We identify three asymmetries between how institutional accountability is structured and how people perceive the process: epistemic asymmetry in access to decision logic, jurisdictional asymmetry in exposure shaped by geopolitical positioning, and temporal--relational asymmetry in how waiting and uncertainty are experienced. We emphasize why it is important to shift attention from institutional design to the uneven distribution of experiences with public-sector algorithmic governance. Together, these contributions demonstrate how algorithmic governance systems in the context of transnational migration produce structured asymmetries not captured by institutional disclosure frameworks, and how extending ADMAPS can account for those uneven translations of accountability.

Homelessness systems in North America adopt coordinated data-driven approaches to efficiently match support services to clients based on their assessed needs and available resources. AI tools are increasingly being implemented to allocate resources, reduce costs and predict risks in this space. In this study, we conducted an ethnographic case study on the City of Toronto's homelessness system's data practices across different critical points. We show how the City's data practices offer standardized processes for client care but frontline workers also engage in heuristic decision-making in their work to navigate uncertainties, client resistance to sharing information, and resource constraints. From these findings, we show the temporality of client data which constrain the validity of predictive AI models. Additionally, we highlight how the City adopts an iterative and holistic client assessment approach which contrasts to commonly used risk assessment tools in homelessness, providing future directions to design holistic decision-making tools for homelessness.

Abstract Teaching has often been characterized as a “labor of love.” Despite their passion, teachers often find themselves underpaid and unrecognized, leading them to engage in taxing emotional labor. Emotional labor in traditional educational settings is not new. However, teaching as online gig work has become increasingly data-driven and transnational. With the burgeoning popularity of online educational industries in China, U.S. teachers are entering the transitional gig economy to teach students, parents, and educational standards in cross-cultural contexts. Based on 24 semi-structured interviews with U.S. teachers who worked on Chinese gig-education platforms, this paper documents their challenges and how such platforms reconfigure their emotional labor, enabling them to reaffirm their identities as teachers and caregivers and rekindle the passion that gave their lives purpose and meaning. However, these platforms, underpinned by Chinese cultural values and data-driven technologies (e.g., datafication, algorithms, and surveillance) — which we dub transnational emotional computing — demand emergent forms of emotional labor with which participants must contend. This work contributes to a human-centered conceptualization of identity alignment and carries theoretical and design implications for the future of transnational gig platforms, especially for cross-cultural digital knowledge labor.

Informed by the previous HCI and CSCW scholarship on residual mobility -- a concept that transcends mere geographical relocation to encompass socio-cultural and communal disruptions -- this study probes the unique religious and spiritual challenges faced by the Hindu migrants from the Indian subcontinent in Canada. Through interviews with 20 participants, we investigate the role of technology in navigating a diverse religious landscape in professional environments, coping with changing religious materiality, and passing down traditions to the next generation. Our work identifies the community's proactive use of social media and videoconferencing for religious festivals and connection with their religious community. The findings raise several implications for CSCW research on supporting residual mobility experiences of the Hindu migrants, including effective organization of religious event information, virtual support for material aspects of religious rituals, and fostering online environments that enable pluralistic spiritual engagement.

Large language models (LLMs) are the new hot trend being rapidly integrated into products and services—often, in chatbots. LLM-powered chatbots are expected to respond to any number of topics, including topics central to gender identity . In light of rising anti-trans discourse, we examined how two popular LLMs responded to real-world English-language questions about trans identity taken from Quora. We employed reflexive analysis that centered our situated knowledges of the trans community. We found that LLMs return pro-trans responses, even when presented with highly transphobic user prompts. While we also found highly transphobic LLM responses, we found that anti-trans sentiment in LLMs was often subtle, requiring a deep positional understanding from diverse trans stakeholders to interpret. Based on these findings, we recommend diverging from current “value-neutral” approaches that validate transphobia by taking an “all sides” approach. We provide considerations for both the evaluation and design of LLMs that center positional expertise.

While colonization has sociohistorically impacted people’s identities across various dimensions, those colonial values and biases continue to be perpetuated by sociotechnical systems. One category of sociotechnical systems–sentiment analysis tools–can also perpetuate colonial values and bias, yet less attention has been paid to how such tools may be complicit in perpetuating coloniality, although they are often used to guide various practices (e.g., content moderation). In this paper, we explore potential bias in sentiment analysis tools in the context of Bengali communities who have experienced and continue to experience the impacts of colonialism. Drawing on identity categories most impacted by colonialism amongst local Bengali communities, we focused our analytic attention on gender, religion, and nationality. We conducted an algorithmic audit of all sentiment analysis tools for Bengali, available on the Python package index (PyPI) and GitHub. Despite similar semantic content and structure, our analyses showed that in addition to inconsistencies in output from different tools, Bengali sentiment analysis tools exhibit bias between different identity categories and respond differently to different ways of identity expression. Connecting our findings with colonially shaped sociocultural structures of Bengali communities, we discuss the implications of downstream bias of sentiment analysis tools.

While content moderation in online platforms marginalizes users in the Global South at large, users of certain identities are further marginalized. Such users often come from Indigenous ethnic minority groups or identify as women. Through a qualitative study based on 18 semi-structured interviews, this paper explores how such users' experiences with hate speech online in Bangladesh are shaped by their intersectional identities. Through a civics-oriented approach, we examined the spectrum of their legal status, membership, rights, and participation as users of online platforms. Drawing analogies with the concept of citizenship, we develop the concept of usership that offers a user-centered metaphor in studying moderation and platform governance.

Colonialism--the policies and practices wherein a foreign body imposes its ways of life on local communities--has historically impacted how collectives perceive themselves in relation to others. One way colonialism has impacted how people see themselves is through nationalism, where nationalism is often understood through shared language, culture, religion, and geopolitical borders. The way colonialism has shaped people's experiences with nationalism has shaped historical conflicts between members of different nation-states for a long time. While recent social computing research has studied how colonially marginalized people can engage in discourse to decolonize or re-imagine and reclaim themselves and their communities on their own terms--what is less understood is how technology can better support decolonial discourses in an effort to re-imagine nationalism. To understand this phenomenon, this research draws on a semi-structured interview study with YouTubers who make videos about culturally Bengali people whose lives were upended as a product of colonization and are now dispersed across Bangladesh, India, and Pakistan. This research seeks to understand people's motivations and strategies for engaging in video-mediated decolonial discourse in transnational contexts. We discuss how our work demonstrates the potential of the sociomateriality of decolonial discourse online and extends an invitation to foreground complexities of nationalism in social computing research.

Despite significant research on online harm, polarization, public deliberation, and justice, CSCW still lacks a comprehensive understanding of the experiences of religious minorities, particularly in relation to fear, as prominently evident in our study. Gaining faith-sensitive insights into the expression, participation, and inter-religious interactions on social media can contribute to CSCW's literature on online safety and interfaith communication. In pursuit of this goal, we conducted a six-month-long, interview-based study with the Hindu, Buddhist, and Indigenous communities in Bangladesh. Our study draws on an extensive body of research encompassing the spiral of silence, the cultural politics of fear, and communication accommodation to examine how social media use by religious minorities is influenced by fear, which is associated with social conformity, misinformation, stigma, stereotypes, and South Asian postcolonial memory. Moreover, we engage with scholarly perspectives from religious studies, justice, and South Asian violence and offer important critical insights and design lessons for the CSCW literature on public deliberation, justice, and interfaith communication.

While people’s identities can be marginalized through various forces, colonialism is one of the primary ways that continues to influence people’s lives and identities. Colonialism refers to the policies and practices where foreign powers migrate to other lands and alter the social structures, and thus identities, of local populations. What is less understood is how online spaces can support people in the aftermath of colonization in revising, repairing, and strengthening their identities—the process of identity decolonization work. Using trace ethnography beginning on 15 May, 2020 and ending on 15 July, 2020 and drawing on Poka Laenui’s framework of decolonization, we explore how South Asian Bengalis on the platform Bengali Quora (BnQuora) engage in collaborative identity decolonization work to reclaim narrative agency. We discuss how narratives serve to help people bounce back from threat or vulnerability—a concept we dub narrative resilience. We also describe potential implications for future scholarship focused on decolonization that extends multiple ongoing conversations around ICT for development, social justice, decolonial HCI, and identity research within the CHI community.

Nowadays much of human interaction are taking place online and on social networking sites. These platforms often encourage people to use profile pictures as parts of their profiles. To understand digital identity construction across various user communities and to foster inclusive, diverse, and sustainable online interactions among them, it is imperative to understand the perception of different users groups about digital profile pictures (DPP). In order to explore this, we conducted a cross-border analysis on different social media platforms. Based on a quantitative study on more than 500 responses from a two-week survey of social media users from 29 different countries, we observed how people from various demographic groups perceive the importance of DPPs. Our results suggest that the perception is significantly different across some social factors and social media usage behavior.

Researchers heavily use online platforms for collecting trace data, i.e., data capturing user interaction on and with sociotechnical systems. Human-computer interaction scholars have highlighted the role of reflexivity while analyzing such data in the case of marginalized communities. Drawing on sociomaterial perspectives, we highlight how data collection approaches involving lists of search phrases and APIs can embed researchers’ positionality, perspectives, and biases within the datasets. In this note, we reflect on the data collection approaches of two papers that studied the sociohistorically marginalized Bengali communities on the question-and-answer site Bengali Quora. We illustrate how recommendation systems and data labeling workers can be included in the data collection process to democratize and limit bias while broadening and contextualizing the trace datasets for research.

In this paper, we study popular microcelebrities from the Global South to understand their strategies and practices on Facebook. Unlike traditional celebrities who gain their reputation through different types of physical performance, these microcelebrities attain their status by presenting themselves in a favorable way to their online followers. We conducted interviews with 19 microcelebrities from Bangladesh and analyzed our data using actor-network theory (ANT) and Goffman’s dramaturgical analysis (DA) of human interaction. We discuss the complex socio-technical ecosystem of the microcelebrity and the roles of non-human actors, such as platforms and local internet infrastructure along with human actors’ practices. We explain the microcelebrities’ experience with the process of microcelebritification–the process of being a microcelebrity on social media through impression management, and becoming opinion leaders in local sociopolitical discourses as part of their online identity. Our paper contributes to the emerging literature on microcelebrities by highlighting the process viewed in the context of the Global South.

In this paper, we explore how sociotechnical systems support and impede the identity performances and identity expression of communities that have experienced a long history of colonialism, where colonization is the practice through which a foreign power reshapes the social structures and systems of other societies. We conducted a trace ethnography among members of a specific digital platform-Bengali Quora (BnQuora). BnQuora is part of the question and answer (Q&A) platform Quora, where people with this particular ethnolinguistic identity come together to engage in conversations about their identities; identities which were shaped through a long history of colonization in the Global South. In drawing on a conceptual framework that brings together identity performativity, governance, content moderation, and surveillance, we find that the sociotechnical mechanisms of governance that mediate people's performances on the BnQuora platform give rise to a kind of platform identity-certain identities are privileged while others are pushed to the margins based on linguistic practices, nationalities, and religious affiliations. We illustrate this through the themes of moderators as prison guards, collective surveillance as enforcing a majority identity, algorithmic coloniality, and staging as self-imprisonment. Finally, we discuss the ways in which governance shapes a platform's identity and can create, strengthen, and reinforce coloniality.

Stigma, a critical challenge for social justice, has not received much attention in ICTD literature. Most existing designs that aim to combat stigma draw on an 'information and awareness' approach that is often inadequate to address stigma's deeper roots. To address this gap, we have conducted an interview and design study in five special needs schools1 in Dhaka, Bangladesh, involving twenty-nine parents and nine teachers. Based on our study, we present how the primary caretakers of children with autism face the stigma associated with autism spectrum disorder (ASD) and how misogyny, economic apprehension, and misinformation generate this stigma. Drawing from a range of scholarly work in sociology and psychology, we demonstrate how those factors are rooted in the colonial history and contemporary social hierarchy of Bangladesh. Based on our participatory design sessions, we introduce and analyze potential design directions and connect our findings to the politics of inclusion and social justice in the context of the developing world.

StackOverflow (SO), the most popular community Q&A site rewards answerers with reputation scores to encourage answers from volunteer participants. However, irrespective of the difficulty of a question, the contributor of an accepted answer is awarded with the same 'reputation' score, which may demotivate an user's additional efforts to answer a difficult question. To facilitate a question difficulty aware rewarding system, this study proposes SOQDE (Stack Overflow Question Difficulty Estimation), a supervised learning based Question difficulty estimation model for the StackOverflow. To design SOQDE, we randomly selected 936 questions from a SO datadump exported during September 2017. Two of the authors independently labeled those questions into three categories (basic, intermediate, or advanced), where conflicting labels were resolved through tie-breaking votes from a third author. We performed an empirical study to determine how the difficulty of a question impacts its outcomes, such as number of votes, resolution time, and number of votes. Our results suggest that the answers of a basic question receive more votes and therefore would generate more reputation points for an answerer. Due to less incentives relative to efforts spent by an answerer, intermediate and advanced questions encounter significantly more delays than the basic questions, which further validates the need of a model like SOQDE. To build our model, we have identified textual and contextual features of a question and divided them into two categories-pre-hoc and post-hoc features. We observed a model based on Random Forest achieving the highest mean accuracy (67.6%), using only answer-independent pre-hoc features. Accommodating answer-dependent post-hoc features, we were able to improve the mean accuracy of our model to 75.2%.