Search Results (437)

The rapid growth of social media has intensified the spread of hate speech targeting individuals based on race, gender, religion, and sexual orientation, creating significant technical, social, and ethical challenges. To address this, we propose a comprehensive English–Norwegian framework that unifies multi-class, multi-label, and multi-task learning for hate speech detection while embedding fairness, transparency, and accountability. The framework comprises six phases, including ethical system design, English–Norwegian data collection, bias analysis and mitigation, explainable AI, and stakeholder validation. A 14K-instance English–Norwegian dataset was constructed from public social media posts and enriched with identity-specific attributes (religion, race, gender, and sexual orientation) using human-in-the-loop and model-assisted annotation. The modeling approach applies multi-label binary fine-tuning to detect potentially hateful, offensive, aggressive, and neutral content, alongside multi-task learning for mutually exclusive demographic attributes. To promote ethical AI, identity-related prediction sensitivity is addressed through synonym-based augmentation, identity-conditioned label-contrast augmentation, and balanced loss re-weighting. Transparency is ensured using LIME to provide token-level explanations of model decisions. On the combined held-out evaluation partition, NorBERT-small achieved a macro F1-score of 0.92 and multi-label subset accuracy of 0.94 for toxicity classification. Language-stratified analysis further yielded macro F1-scores of 0.92 for English-only, 0.89 for Norwegian-only, and 0.87 for English–Norwegian code-switched text, providing a more detailed assessment of performance across the different language conditions, and high aggregate F1-scores for social-attribute extraction; however, these aggregate results should be interpreted cautiously because several social-attribute classes are sparsely represented. Compared with the baseline model, the mitigation stage reduces prediction sensitivity to many of the evaluated identity substitutions, although residual disparities remain for some identity pairs. The Machine Learning (ML) and Deep Learning (DL) models are included as conventional reference baselines. Because the evaluated model families employ different architectures and task-specific training procedures, the cross-family results are interpreted descriptively rather than as evidence of the intrinsic superiority of any particular model family. The reported findings are specific to the evaluated corpus; cross-platform, cross-domain, and external-dataset generalization were not evaluated and are therefore not claimed. Full article

Artificial Intelligence (AI) and Media and Information Literacy (MIL) are recognised as vital components of adult secondary education. However, we know little about how adult learners engage with AI in authentic educational contexts, given that AI adoption is a complex sociotechnical process. In this study, AI for MIL (AI/MIL) refers to AI-supported tools that guide learners in interpreting and engaging with digital media and information. We report findings from an ethnographic study involving eight adult students in Greek secondary education. Informed by the Unified Theory of Acceptance and Use of Technology (UTAUT), we followed participants over an eight-month educational intervention through participant observation, interviews, and field notes. Our data show that learners moved from anxiety toward more agentic forms of AI engagement. This ethnographic reinterpretation of UTAUT highlights an interplay among technological, emotional, and sociocultural factors that shape AI/MIL adoption. Beyond traditional technology acceptance models, we underscore negotiated trust, critical AI literacy, ethical awareness, and human agency within AI-mediated information ecosystems. These findings have implications for designing participatory, transparent, and ethically grounded AI initiatives in educational settings. Full article

Amid rising authoritarianism, the criminalization of dissent, and the erasure of indigenous expertise from policy discourse, this paper examines community-engaged research as critical public scholarship through the Deccan Development Society’s (DDS) four-decade-plus collaboration with approximately 5000 Dalit and indigenous women farmers in Telangana, India. Against a backdrop of shrinking civic space and technocratic development agendas that delegitimize marginalized knowledge, DDS women’s sanghams have consolidated village-level assemblies into organs of local governance that claim autonomy over land, seed, food, and knowledge—positioning marginalized women as producers of evidence and theory rather than mere data sources. Drawing on complexity theory and participatory action research (PAR), this paper theorizes a community-based inquiry model where marginalized women operate as co-researchers, knowledge producers, and media creators within an emergent system characterized by distributed agency and non-hierarchical information flows. The paper interrogates key methodological, ethical, and political dynamics of critical community-based inquiry for justice and animates the concept of “autonomy as counter-policy” as a contribution to engaged scholarship and epistemic justice. Full article

This study analyses how Spanish digital news outlets covered the arrival of artificial intelligence in journalism and the social impact they attributed to it. That general objective is pursued through four specific research objectives rather than through hypotheses in a descriptive and exploratory design that traces the emergence and distribution of media attention, identifies dominant and omitted issue-specific frames, and assesses their association with journalistic variables. A longitudinal quantitative content analysis covers 242 news items from ten leading Spanish digital outlets, coded in their entirety rather than sampled, across the three phases of AI’s public visibility between 2022 and 2024. Coverage follows a compressed attention cycle in which enthusiasm and alarm unfold simultaneously rather than consecutively. Dominant frames centre on generative AI in the processing phase, credibility, ethics in the use of data, legal responsibility and information quality, and timeliness and accuracy as its main advantage and disadvantage. Systematically omitted are the effects of AI on the normative functions of journalism. The main source is the variable most closely associated with these patterns: expert sources broaden the range of frames, whereas the overrepresented media-sector actors narrow it. Spanish digital media thus construct AI as an operational rather than a democratic issue. Full article

Artificial intelligence (AI), particularly generative artificial intelligence (GenAI), is influencing journalism and education by transforming professional practices, information production, and media literacy requirements. While generative AI improves efficiency in content creation and educational preparation, it also raises concerns regarding accuracy, credibility, verification, and ethical responsibility. This study examines how generative AI is experienced by journalists and teachers in primary schools in Croatia within the contemporary information ecosystem. Empirical data were collected through semi-structured interviews with ten participants, including five professional journalists and five teachers, and analysed using reflexive thematic analysis. The analysis identified a shared efficiency–verification paradox across both professional groups. Although generative AI reduces the time required for routine cognitive tasks, it simultaneously increases the need for verification, critical evaluation, and ethical judgement. Participants also emphasised the growing importance of AI literacy, critical thinking, and ethical competencies for navigating AI-generated content in professional and educational contexts. The findings suggest that generative AI does not necessarily replace professional expertise but instead redistributes professional responsibilities towards activities requiring human judgement and accountability. This study contributes to a broader understanding of how AI influences interconnected information professions and highlights the need to strengthen competencies required for responsible engagement with AI-generated information. Full article

As professionals increasingly encounter artificial intelligence (AI) in their workplaces, questions around adoption barriers, ethical concerns, and job security have grown in prominence. This paper reports an exploratory cross-sectional descriptive survey of self-reported perceptions among a convenience sample of 324 UK-based industry professionals. The study does not identify determinants, predictors, or causes of AI adoption. The survey examined eight binary items covering daily personal AI use, perceived strategic importance, cost as a barrier, ethical concern, perceived income change, subjective job-displacement anxiety, perceived work-performance change, and preference for conversational AI tools over traditional search engines. All eight items were completed by all 324 eligible respondents. Using Wald 95% confidence intervals, 47.2% (95% CI 41.8–52.7) reported using AI in their daily jobs, while 66.4% (95% CI 61.2–71.5) perceived AI as important for remaining competitive. Cost was identified as a barrier by 76.5% (95% CI 71.9–81.2), and 65.7% (95% CI 60.6–70.9) reported concern about ethical implications. These two figures are separate aggregate proportions and are not treated as an individual-level behavioural gap. Only 37.0% (95% CI 31.8–42.3) reported an income increase associated with AI at work, whereas 71.9% (95% CI 67.0–76.8) reported improved work performance. These two items were separate self-report questions; no respondent-level association was tested, and neither item measured objective productivity or pay. Job-automation concern was reported by 48.5% (95% CI 43.0–53.9). The same share as the performance item, 71.9% (95% CI 67.0–76.8) preferred ChatGPT-like tools over traditional search engines. The sample was recruited through professional networks, email, social media, and organisational mailing lists and is likely to over-represent professionals already interested in digital tools. Findings are therefore presented as descriptive perceptions from this respondent pool and are not generalised to UK professionals as a whole. Full article

Wild-food foraging is booming across the Global North, but its meaning—and its value for plant conservation—differs sharply between settings. We compare three case studies: Denmark, an intensively cultivated country with a book- and app-led revival; Southern Scotland, a post-industrial landscape whose foragers are largely incomers with a media-sourced repertoire; and Tel Aviv–Jaffa (Israel), where Nordic-inspired “neo-foraging” coexists with an eroding Arab–Palestinian oral tradition. Pooling three independently collected ethnobotanical datasets (2012–2021; Denmark: questionnaires n = 27 and interviews n = 10; Southern Scotland: questionnaires n = 12 and interviews n = 9; Tel Aviv–Jaffa: interviews only, n = 10), we record 153 wild-food taxa—136 plants (including three macroalgae) and 17 fungi—in 51 plant families. Read through the Latvian (Riga) sociology of foraging, all three cases sit on the delocalised, “lifestyle” side of the forager spectrum. Yet, all three converge on one under-exploited fact: foraging can train ordinary urban people to see, name, track and value wild plants. The apparent contradiction between gathering plants and conserving them, we argue, is resolved by the target flora rather than by the practice itself: harvest risk is concentrated in a small minority of rare, range-edge and habitat-specialist taxa, whereas plant awareness is built through repeated encounters with the commonest species—precisely those whose gathering is demographically trivial. Where foragers anchor their teaching in this ordinary, ubiquitous flora, their role is not merely compatible with plant conservation but quintessential to it. Foragers are, thus, an underused constituency of citizen educators capable of countering plant awareness disparity (also termed “plant unawareness”), provided their enthusiasm is coupled to a stewardship ethic, and we set out how conservation actors might mobilise rather than merely regulate this potential. Full article

In cryptocurrency markets, credibility may be created before it can be verified. Promotional signals from influencers, online communities, exchanges, and project actors can create an impression of legitimacy before consumers are able to independently assess the quality, risks, or underlying value of the asset. This systematic review examines how promotional activity contributes to this condition and how the literature connects it with ethical concerns, consumer consequences, and governance. Following PRISMA 2020, 54 empirical studies published between 2019 and 2025 were identified through Scopus and analyzed using thematic synthesis. The findings indicate that promotion is distributed across social media, influencers, communities, exchanges, and project actors, making the boundary between marketing, personal opinion, and financial advice difficult to maintain. Ethical problems arise not only from fraud, but also from selective information, hidden incentives, artificial attention, and market signals that give uncertain assets an appearance of legitimacy. Consumer vulnerability is similarly situational. Knowledge and experience may offer some protection, but trust, technological complexity, social influence, and speculative expectations continue to shape judgement. Regulatory and educational responses remain fragmented because responsibility is dispersed across actors and jurisdictions. The review connects these relationships through an ethical marketing cycle that conceptually organizes how ethical concerns may develop across communication, market activity, consumer interpretation, and governance. This perspective extends ethical marketing beyond the accuracy of individual claims and locates responsibility within the structures through which cryptocurrency credibility is produced. The review is limited by its reliance on Scopus and the absence of a formal risk-of-bias assessment of the included studies. Full article

This article presents a practice-oriented methodological proposal for designing inclusive multimodal learning content with Generative AI (GAI) in migrant adult literacy. It does not report an experimental intervention or a statistical evaluation of learning outcomes. Its contribution lies in formalizing a context-sensitive design pathway for early preA1–A2 literacy and language-learning provision in Italian CPIA settings, where learner profiles are highly heterogeneous, attendance may be discontinuous, and written language is both a learning goal and a barrier to participation. Unlike generic AI-supported instructional design frameworks, the proposed approach starts from recurrent communicative needs in adult migrant education and translates them into short, modular and reusable learning artifacts that coordinate textual, visual, audio-oral and interactive layers. The framework distinguishes multimodal design, understood as the pedagogical coordination of different semiotic modes, from the mere use of multiple media. It also integrates accessibility as a set of concrete design criteria, including linguistic readability, visual clarity, audio quality, layout, font size, contrast, cognitive load and usability in print or mobile formats. The article outlines a sequence of design operations: mapping learner profiles, selecting situated communicative scenarios, generating and revising textual material, developing visual and audio scaffolds, structuring guided interaction, and applying pedagogical, cultural and ethical review. An illustrative micro-unit on asking for information at a municipal office shows how this pathway can support dialog, visual glossary, audio practice, role-play and formative assessment. The proposal is intended for CPIA educators, adult literacy professionals, instructional designers and researchers in multimedia learning and educational technology. Its educational implication is that GAI can support inclusive material design only when its outputs are treated as provisional resources to be selected, adapted and validated through human pedagogical judgment. Full article

Public social-media content often contains self-disclosed personal attributes that appear low-risk in isolation but become privacy-relevant when linked across posts, platform accounts, or user-level traces. Existing research has advanced privacy-sensitive content detection, de-anonymization analysis, social-media research ethics, and privacy-compliance workflows; however, limited work operationalizes how personal-data disclosures combine structurally and how these structures can be translated into auditable governance actions. This paper proposes SEM-PDPL, a computational, privacy-law-informed risk-assessment framework for modeling public social-media exposure as semantic exposure graphs and mapping graph patterns to controls aligned with the United Arab Emirates Personal Data Protection Law (PDPL) and compatible with GDPR principles. SEM-PDPL combines governance scoping; a PDPL-informed disclosure taxonomy; hybrid extraction using rule-based methods; named-entity recognition; fine-tuned BERT; and schema-constrained large language model annotation, followed by graph construction at post, platform, corpus, and persona levels. The framework is evaluated on a synthetic multi-platform corpus of 1095 posts generated for 150 personas across 290 platform accounts. Results show that, within this controlled synthetic corpus, fine-tuned BERT provides the strongest extraction performance among six evaluated methods, achieving a macro-F1 of 0.975. Graph analysis shows that exposure density increases with aggregation, rising from 0.275 at post level to 1.000 at corpus level, and from 0.859 at platform level to 0.967 at persona level. Across all graph resolutions, quasi-identifiers emerge as the dominant weighted-degree and betweenness node, indicating that ordinary location, employer, school, and demographic cues often function as bridges connecting sensitive categories such as health and biometric data to identifying information. These findings indicate that, within this controlled corpus, privacy risk in public social-media data is not only attribute-based but also structurally graph-shaped. SEM-PDPL contributes an explainable and reproducible framework for identifying exposure hubs, sensitive bridges, and aggregation risks before applying masking, minimization, exclusion, retention, or review controls. The framework does not automate legal compliance; rather, it provides evidence-based decision support for privacy-aware social-media analytics. Full article

Non-human animals are systematically excluded from public narratives, rarely recognized as sentient subjects, and instead framed within anthropocentric narratives of production and consumption; efforts to challenge this pattern and make the more-than-human world visible often encounter editorial resistance and structural constraints. This study examines how anti-speciesist journalists in Spanish written media navigate such resistance to make non-human animals legitimate subjects of news. Drawing on semi-structured interviews with nine self-identified anti-speciesist journalists, analysed through reflexive thematic analysis, the study identifies strategies organized across four non-mutually exclusive categories, namely, discursive, relational, framing, and structural, ranging from self-censorship and strategic language use to coalition-building and the creation of independent platforms. Findings reveal a key distinction between visibility strategies, which succeed in securing publication, and root-level strategies, which challenge the carnist and anthropocentric assumptions structuring newsworthiness itself. These practices reveal the pragmatism, creativity, and resilience of journalists working under structural, ideological, and normative constraints, while also highlighting the limits of anti-speciesist reporting within mainstream media. The study contributes to Critical Animal Media Studies, environmental communication, and journalism research by illuminating the intersection of ethical advocacy, professional strategy, and counter-hegemonic practice and offers recommendations for expanding coverage of the more-than-human world across ideological, editorial, and structural divides. Full article

Britain’s historical spectacles of animal performance, from Victorian circuses and travelling menageries to early visual media, did more than entertain, they shaped how humans imagined, controlled, and valued nonhuman life. Animals were transformed into objects of fascination and domination, while hunting and poaching practices reinforced imperial authority over wildlife. These performances codified visual and moral frameworks that positioned Britain as arbiter of “wildness,” embedding hierarchies of power and ethical order into the cultural imagination. This paper draws on performance studies, postcolonial ecocriticism, and visual anthropology to trace how circus rings, menageries, and early animal films laid aesthetic and ideological foundations that resonate in contemporary British environmental art and regeneration discourse. By historicising these visual legacies, it reveals how dominion, spectacle, and exploitation persist as underlying structures in Britain’s engagement with nature. Yet contemporary environmental arts are not merely reactive, they reinterpret these histories, transforming frameworks of control into imaginative possibilities for ethical, regenerative, and multispecies representation. By linking past and present, this study demonstrates that understanding the visual politics of historical animal performance is essential to imagining sustainable and inclusive futures for human and nonhuman life in Britain. Full article

Gambling disorder is a public health concern, but research on the early detection of gambling-related harm is limited by the scarcity of ethically shareable online conversations. This study presents a framework for generating an entirely synthetic dataset for natural language processing research on gambling-related behavioral signals. Behavioral dimensions from the Online Gambling Disorder Questionnaire were used to construct 2100 risk-aligned user profiles and generate X/Twitter-style monologues and multi-user threads across gambling and non-gambling subtopics. The dataset contains 35,346 monologues and 1716 threads. The generated text was evaluated using automatic metrics, manual target-consistency assessment, and downstream classification under user-disjoint and parent-topic-disjoint protocols. Automatic and manual evaluations indicated acceptable linguistic quality and consistency with predefined behavioral targets. In the stricter parent-topic-disjoint setting, logistic regression achieved a macro-$F_1$ of 0.884 for binary gambling-content detection, whereas four-level risk prediction remained more difficult. These results show that the dataset contains learnable signals for computational detection of gambling-related content without exposing real-user data. However, the dataset was not externally validated against authentic social-media conversations and is not intended for clinical diagnosis. The resulting resource is designed to support reproducible research on gambling disorder and mental health while preserving user privacy. Full article

Fetal bovine serum (FBS) is commonly added to cell culture media, but for cultivated meat production to become truly sustainable, ethical and affordable, an alternative to it is necessary. The purpose of this study was to demonstrate how myoblasts could be cultivated in reduced-serum media using animal-free scaffolds. Novel chitosan/carrageenan hydrogels were developed and conjugated with peptides obtained by digestion of C-phycocyanin from cyanobacteria Arthrospira platensis, their physico-chemical properties were assessed, and their ability to serve as scaffolds for myoblast cultivation in reduced-serum media was examined. These hydrogels were found to possess suitable physical characteristics to serve as scaffolds for cell culture, and it was demonstrated that their use made quail muscle clone 7 (QM7) cell cultivation possible in reduced-FBS media. These hydrogels, therefore, are a promising candidate for use as scaffolds in future cultivated meat production studies. Full article

This systematic review investigates the application of artificial intelligence (AI) and machine learning (ML) in journalism and media practice from 2020 to 2026. Following PRISMA guidelines, we analysed 121 peer-reviewed articles from Scopus and Web of Science using a multi-method approach that combined qualitative thematic analysis, structural topic modelling (STM), and bibliometric network analysis. Four primary research domains emerged: news production and automation (38.0%), audience perception and content analysis (24.8%), ethical and legal considerations (19.8%), and meta-research and implementation studies (17.4%). Publication output accelerated sharply from 2023 onward, driven by the emergence of large language models and generative AI. The STM analysis confirmed the four-domain structure and revealed that legal-regulatory vocabulary pervades the literature across all categories, indicating a field-wide preoccupation with the institutional implications of AI. Key findings demonstrate that successful AI adoption depends on workflow redesign and human–machine collaboration rather than full automation; that audience evaluations of AI-generated content vary significantly across cultural contexts, with the machine heuristic—originating from Sundar’s MAIN model—playing a central mediating role; and that copyright frameworks for AI-generated news remain contested across jurisdictions. This study develops an integrated theoretical framework that maps directional relationships among the four research domains, identifies cultural context and disclosure practices as key moderators, and generates testable propositions for future investigation. Full article