Computers and Society Papers

ALT As society, ethics, and professional practice become increasingly integrated across the computer science curriculum, technical courses need practical ways to support ethical reasoning. Ethical essay assignments offer one approach, but little is known about how students use GenAI when writing such essays. This experience report examines GenAI use across three years of ethical essay assignments in a large first-year computer systems course. Across six post-essay surveys from 2024 to 2026, we analyzed 1,894 responses, linked survey data with essay grades in 2025 and 2026, and introduced a scaffolded responsible-use workflow in 2026. Under voluntary-use policies, AI adoption remained steady at around two-thirds of students, rising above 90% when use was scaffolded. Students reported mainly using AI for help with grammar and style, while tool use diversified beyond ChatGPT in 2026. Grade-linked analyses showed that AI use did not independently predict essay quality, while marker evaluations

ALT The explosion of generative AI systems and concerns surrounding their rapidly scaling capabilities has led to an adversarial cycle between increasingly narrow content restrictions ("guardrails") and increasingly sophisticated structured prompting techniques ("jailbreaks"). Technological frameworks surrounding such jailbreaking remain largely focused on the system's output: whether harmful or disallowed content is ultimately produced for the user. However, this approach, which draws largely from classical cybersecurity, does not address the uniquely relational dimension introduced by generative AI: in many cases the user is not just accessing a technology via an exploit, but is in conversation with it as part of the exploit. Here we present an alternative lens which centers the interaction rather than the output. Drawing on interpersonal violence scholarship, we present a novel framework of coercive control categories with a particular focus on aspects which are embodied versus disembod

ALT AI evaluation is increasingly central to AI safety and governance, yet taking part in an evaluation does not by itself confer authority over its design, its execution or the resources it depends on. This paper examines how infrastructure arrangements distribute that authority, and so shape who can evaluate AI systems independently. Drawing on Science and Technology Studies, political economy and applied evaluation standards, we introduce PAAI (Power, Access, Adequacy, Independence), a framework for examining who controls evaluation infrastructure, the terms on which it is available, whether it is adequate to the evaluation's purpose and what autonomy it supports. We apply PAAI to twelve African AI evaluation projects (2022-2025) and to the documented roles of all ten member institutes in the three published joint testing exercises of the international network of AI Safety Institutes. We identify three recurring configurations: extractive collaboration, dependent innovation and constrai

ALT Over the past three decades, technological advances have assisted individuals to gain access to legal information, enhance the efficiency of legal services, and streamline dispute resolution. More recently, the proliferation of generative artificial intelligence has reinforced this emphasis on firm-level productivity, with commercial legal practice increasingly leveraging artificial intelligence tools to automate document creation and optimise case management. While commercial technology focuses on productivity, community organisations are exploring how artificial intelligence can level the playing field for marginalised communities. Far less attention has been given to free, direct-to-consumer legal technologies, as these models are often unprofitable. Consequently, justice technology scholarship has overlooked a distinct class of technology: digital reporting platforms that enable people to record their experiences of harm, while simultaneously aggregating those accounts into a body

ALT Language-generating AI systems are no longer only objects of linguistic interaction: their outputs increasingly re-enter the communicative and institutional practices from which their training material was drawn. Drawing on social theory, social epistemology, and the philosophy of language, this article develops a sociotechnical account of that re-entry. We introduce recursive semantic normalization: the process through which machine-generated formulations, once recurrently taken up, evaluated, reused, and institutionally retained, participate in the reproduction and transformation of semantic regularities. Under the heading of technohybridism, we argue that the relevant unit of analysis is the hybrid semantic system, a recursively coupled configuration of human practices, computational generation, and institutional uptake. First, we distinguish the concept from neighbouring accounts, including model collapse, looping effects, machine habitus, machine culture, and the epistemology of A

ALT tok is a plain-text notation for Business Process Model and Notation (BPMN) 2.0.2 process models and Decision Model and Notation (DMN) 1.5 decisions. One engine checks a tok document against both standards, compiles the document to XML with an automatic diagram, and serves language-model applications through the Model Context Protocol. The paper classifies the notation as an alternative concrete syntax of the BPMN and DMN metamodels and reports one study. A tok model takes 2.2 times fewer tokens than the smallest BPMN XML of 893 files exported by 37 tools and 7.7 times fewer than the files as exported. The checker signals 98.2% of 1,411 injected faults, against 85.8% for the XML Schema with bpmnlint. Four language models new to tok write valid tok from a one-page reference for 85.7 to 100% of 87 process descriptions from two datasets and valid BPMN XML for every description, and the strongest model writes tok with 25% and 19% fewer tokens than BPMN XML. As agents with a lean skill and

ALT The tension between news personalization and information diversity has traditionally been understood as a matter of different content reaching different people, with diversity evaluated by the range of sources, topics, and perspectives to which audiences are exposed. Generative AI outdates this understanding, since it transforms personalization in two ways: how journalists adapt news to audiences and how the AI tools journalists use personalize content production to their own preferences. Drawing on 16 semi-structured interviews with media professionals across public service, commercial, and state-affiliated news settings, we examine how these practices reshape editorial judgment and the conditions under which news diversity is produced. We find that participants negotiate AI integration by maintaining authority over meaning while delegating aspects of content presentation. Yet the tools they adapt for their own work also mediate which sources and perspectives become available, introdu

ALT Advances in generative AI have driven a surge in synthetic non-consensual intimate imagery (NCII) and child sexual abuse material (CSAM), prompting a series of legal reforms. These reforms rely on legal definitions where technical operationalization is ambiguous, such as whether a depicted person is identifiable, whether a model is capable of or intended for abuse generation, and whether an abusive image can be traced to the model that produced it. Drawing on the law of England and Wales and the United States, we examine the relevant law alongside the technical research needed to operationalize it. From this, we develop a research agenda of open problems at this technical-legal interface, spanning the interpretation of existing law, the prevention of harm before an AI tool is deployed, and enforcement of offenses. These are problems the technical community is well placed to solve, and where progress is vital to real-world protection.

ALT The shift from traditional to generative search has fundamentally transformed online product discovery. Because conversational recommendations synthesize choices into a few direct answers, subtle systemic preferences toward specific providers can exert profound commercial impacts. In this work, we audit this phenomenon through two distinct dimensions: (i) brand popularity bias, the systematic preference for market-dominant entities over long-tail alternatives, and (ii) brand familiarity bias, the ranking advantage an entity encoded in an LLM's parametric memory holds over an unfamiliar competitor. First, evaluating commercial queries across major engines, we observe persistent brand popularity bias: dominant brands in the top quintile are cited 1.4x to 2.8x more frequently than the average brand within a sample of 1,176 top market players, already present in parametric memory that external web retrieval fails to mitigate. Second, through controlled in-context experiments holding produc

ALT North Korean defectors residing in China face persistent uncertainty regarding digital security and privacy. Through semi-structured interviews with 19 defectors, we examine how individuals understand digital threats and adapt their security behaviors. We identify three interrelated conditions: transnational surveillance pressure from North Korea, identity-linked digital infrastructure and intermediary dependency in China, and threat reasoning shaped by prior surveillance schemas. Participants developed uneven protective practices and followed two trajectories: adaptive habituation driven by knowledge acquisition and erosive habituation driven by the prolonged absence of adverse events. Our findings show how compound threats, attribution uncertainty, and constrained agency shape security reasoning among undocumented users.

ALT Generative AI tools are now widely available to students, but little is known about how team-based project codebases have changed since they appeared. This matters when project courses use repository activity as evidence of progress and learning. Existing studies have largely examined self-reports, controlled tasks, or short individual assignments, leaving open questions around how student codebases evolve in authentic project settings. We present a ten-semester analysis of N=260 capstone team repositories from 2021 to 2025, spanning pre- and post-AI cohorts. We examine three dimensions: code size (lines of code), code characteristics (comment density, error-handling density, and cyclomatic complexity), and development behaviour (commit volume, edit size, and commit frequency). We find that end-of-semester codebases roughly triple, from about 3.4 KLOC in 2021–2023 to about 10 KLOC in 2024–2025. Code characteristics show a clear decoupling: cyclomatic complexity and error-handling densi

ALT In the years following the public release of ChatGPT in late 2022, more than half of U.S. states have published formal guidelines around AI usage in K-12 education, creating a record of largely independent state education departments responding to a shared technological disruption. In this work, we analyze publicly available AI usage guidance documents from 35 states in the U.S. and Puerto Rico. We compare the guidance in these documents across various dimensions, including pedagogy and instruction, audience, actionability, workforce and career readiness, responsible use and equity, and alignment with state-level demographics and administrative capacity. Finally, we investigate who authored these documents and their cited sources, identifying epistemic foundations underlying their claims. We find that guidance priorities converge on topics anchored in laws such as data privacy and academic integrity, each addressed in 97% of documents. While most documents mention equity of access (86%

ALT LLMs can predict human survey responses with nontrivial accuracy. What this means for the future of social surveys remains hotly contested. Optimists anticipate a coming era of synthetic respondents; skeptics see a fragile and overhyped substitute. This article advocates a position: AI predictions cannot replace surveys, but they can augment them. Using prediction-powered inference (PPI), a statistical framework rooted in survey methodology and known in sociology as "mixed subjects design", the study tests whether AI predictions can improve the statistical efficiency of social surveys when combined with census-scale auxiliary information. Applying the framework to 83 social outcomes in the General Social Survey (GSS) and National Health Interview Survey (NHIS), with American Community Survey (ACS) being the auxiliary file, we find that supervised predictors yield effective-sample-size gains of about 7-12 percent when estimating the mean, while LLM gains are smaller but positive.

ALT Responsible AI scholarship has found that ethical principles are frequently not reflected in AI design and development. Conceptualised as a `gap' between principle and practice, this is framed as a problem of implementation, awareness or compliance, and by implication, a deficit of systems or individual behaviour. This paper sets out an alternative explanation. Drawing on fourteen semi-structured interviews with actors designing and governing AI-enabled anti-trafficking systems across law enforcement, commercial, research, and survivor-support contexts, it argues that such apparent gaps emerge from the varied and complex institutional and professional conditions through which actors come to recognise some issues, but not others, as ethical problems requiring action. We show that epistemic choices are intertwined with ethical judgements, and trace four interconnected moments—perception, judgement, situated action and socio-technical materialisation—through which concerns become intellig

ALT Public bodies increasingly rely on AI-produced evidence, including classifications, summaries and assessments. These outputs can inform policy decisions and service delivery, making their reliability a matter of public accountability. Establishing whether this evidence can be relied upon requires testing its outputs and examining how organisations decide whether those tests are sufficient. We combine technical evaluation with analysis of AI governance requirements, informed by measurement theory and management-based regulation. We compare eight GPT-4o configurations classifying 1,200 UK Prime Minister's Questions into 21 policy topics, against human-coded labels and across unchanged repeats. Accuracy ranges from 68.2% to 70.3%, yet 24.3% of questions receive different labels, with changes in topic shares and rankings. Configurations agree more with one another than with human coding: agreement does not guarantee correctness. Under the implemented score-extraction procedure, the proport

ALT Large language models (LLMs) can be prompted to express different personality profiles, but it remains unclear how reliably they do so or whether personality affects their persuasiveness. We investigate these questions in two evaluations. First, we assess Big Five personality prompting using the IPIP-NEO-120, examining reproducibility, trait controllability, cross-trait effects, and unconditioned model profiles. Second, we conduct a within-subjects user study (N=32) in which participants discuss moral dilemmas with LLM agents whose personalities are aligned, mismatched, neutral, or unconditioned. We find that prompted profiles can be reproducible and responsive to requested traits, although controllability varies across models and changes to one trait can affect others. Unconditioned models also exhibit systematic personality profiles rather than neutral baselines. In the user study, personality alignment did not significantly affect perceived persuasiveness, stance switching, or confi

ALT Researchers routinely infer gender, race, and ethnicity from names, yet the standard tools used for this task misclassify the groups that diversity research most needs to measure. Large language models (LLMs) have been proposed as a remedy because they label diverse names accurately. We ask whether that accuracy carries over to the estimates researchers report. Across 10,000 Chinese names, 19,998 North Carolina voters with self-reported race, and 300 political science authors with human reference codes, we compare three LLMs with seven traditional tools and link the voter sample to party registration and age. Selective labeling explains part of the LLMs' accuracy advantage. Claude Haiku labels half of the voters, and a traditional tool limited to its most confident half is more accurate. Hard labels from every tool understate the White–Black gap in Democratic registration by 36% to 65%, while BIRDiE estimation on BISG posteriors comes within 0.03 of it. Haiku's missing labels select on

ALT Modern societies and institutions depend on an invisible layer of open digital infrastructure: foundational, openly accessible digital technologies that support a broad range of social and economic activities. Open source software is a core component of this infrastructure, underpinning everyday technologies and critical systems in various critical sectors. As open source software becomes increasingly recognised, funded, and governed as open digital infrastructure, policymakers face growing demands for accountability, transparency, and credible evidence of impact. However, methods for evaluating public investment in open source software remain underdeveloped, with no widely accepted methodological standard. This paper describes the Sovereign Tech Agency's transition from qualitative reporting and funding-recipient self-assessment towards a mixed-methods approach that incorporates quantitative, goal-oriented evaluation to inform funding decisions. Drawing on the development of this fram

ALT Volunteer fire brigades rely on fragmented alerting channels and radio plus paper for situational awareness, hindering a shared common operational picture (COP). FireOPS integrates smartphone alerting, navigation, a live geospatial COP, a district resource-load overview and contextual manuals into a human-centric decision-support surface for commanders and responders, running on an edge + cloud architecture whose dual-channel provides source-level redundancy against failure of either the radio or the provincial server feed; delivery to smartphones still depends on the cloud core and mobile networks. This lessons-learned report presents the design rationale, positions FireOPS within emergency-response information systems research, and reports operational findings from a primary log extract of a major regional flood in 2024 and from the 2025 public app release. A feature-matrix comparison with DIVERA 24/7, blaulichtSMS and the LifeRescue prototype identifies COP, decision support and dua

ALT Computer science research on the "alignment" of artificial intelligence (AI) systems treats human values as content to be specified and optimized. Decades of social science scholarship describe values differently, as enacted in ordinary practice, interpreted in social context, and contested across plural orders of evaluation and worth. We argue that the dominant alignment techniques, which we call "thin alignment," convert situated moral or value judgments into one or a few metrics via devices of commensuration such as preference labeling, reward modeling, and consumer ratings. Drawing on philosophical and anthropological accounts of thickness, "thick alignment" instead examines how values take shape, inform the development of AI systems, and become subject to contestation over whose interests they serve. We describe two forms of misalignment in seemingly aligned algorithmic systems that highlight thin alignment: "Halcyon misalignment," which presumes a prior, just equilibrium of techn