Conference Agenda
| Session | ||
31 SES 08 B: AI
Paper Session | ||
| Presentations | ||
31. LEd – Network on Language and Education
Paper GenAI as Sociomaterial Actor in Undergraduate Students’ Academic Writing 1: Mälardalen University, Sweden; 2: Åbo Akademi, Finland; 3: Helsingfors university, Finland Presenting Author:This paper explores GenAI as a relational actor within university students’ academic writing. The study is grounded in students’ reported accounts of how various human and non-human actors participate in and shape their academic writing. The study focuses specifically on how students position GenAI within the relational networks of academic writing that they articulate. The paper engages with a relational understanding of academic writing as an emergent, distributed, and materially mediated practice, thinking with relational ontological perspectives (e.g., Barad, 2007; Brandt & Clinton, 2002; Latour, 2005). From such perspective, writing is not solely the product of an individual writer but constituted through relations among multiple human and non-human actors. GenAI is thus not an autonomous or self-contained entity but rather emergent through its relations with the broader material, institutional, and ideological arrangements that constitute students’ academic work. Recent international and Nordic surveys underscore the importance of exploring students’ use of GenAI. A global survey by the Digital Education Council (2024) reports that 86% of students use AI in their studies, and a UK-wide survey by HEPI (Freeman, 2025) shows that 92% of undergraduate students draw on AI tools for academic work. In Sweden, a national survey of nearly 6,000 university students finds that almost all students are familiar with ChatGPT and more than one third use it regularly for learning purposes (Malmström et al. 2023). Research have also pointed to the various ways on which GenAI is incorporated into student work, for instance in relation to idea generation in writing (e.g. Barrett & Pack, 2023; Kim et al., 2025; La Scala et al., 2025; Tsufim & Pomerleau, 2024). Given the rapid pace of development and adoption of these technologies, it is reasonable to assume that the presence of generative AI in students’ academic work continues to grow. These trends form an important backdrop for the present study. Against this backdrop, the present study foregrounds undergraduate students’ accounts of their academic writing, generated during workshops embedded within their regular course activities. These workshops provided access to students situated reflections on how their academic writing unfolds. Rather than treating GenAI as an isolated object of study, we adopt a holistic perspective on these reflections, examining how students describe the constellation of actors and relations that shape their writing. This approach enables an examination of how GenAI becomes embedded within, and interacts across, the broader networks that constitute students’ writing practices. At the same time, the approach acknowledges that GenAI represents only one of multiple actors involved in shaping students’ academic writing. Accordingly, the study explores: (1) How GenAI becomes positioned as an actor in students’ reflections on academic writing and (2) How this positioning, in turn, shape and reshapes the relations that constitute students’ academic writing networks. Methodology, Methods, Research Instruments or Sources Used This study adopts an actor–network theory (ANT) informed methodological approach, engaging in what is often described as “doing ANT,” a simultaneous theoretical sensibility and analytic practice (Ahn, 2015). Methodologically, this approach involves tracing how relations are articulated, maintained, and transformed within sociomaterial networks. In this study, we trace generative AI retrospectively through students’ own accounts of their academic writing, focusing on how students describe the actors that are involved in these networks and the emergence of GenAI in these relations (Gunnarsson & Bodén, 2021). Rather than observing writing processes directly, the study focuses on students’ reported experiences and descriptions, treating these accounts as sites where actors and relations are made visible. The study is part of the CO‑WRITE research project (2025–2027), which investigates collaborative academic writing in higher‑education settings. Empirical material was generated through four workshops (2.5 hours each) with 30 students enrolled in educational, political, and caring sciences at a Finnish university. During the workshops, students were invited to map and discuss the human and non-human actors they considered central to their academic writing processes. They were asked to explain why particular actors became central, how these actors’ gained relevance, and what roles they played in shaping their writing practices. The data consist of audio-recorded conversations and student produced textual and visual mappings. The workshops were designed to not only elicit student responses about what they use in their academic writing, how they use it, and why, but also make the research participation valuable for the students (cf. Kara, 2015). As such, the workshops had both a pedagogical and a research-based emphasis, with the intention that something of value might emerge not only for the research but also for the students themselves. Our analytic strategy involved tracing the actors described by students. Within this analysis, ANT offers a flat ontology that enables an examination of how actors, human and non-human, take shape through relations rather than through predetermined qualities (Latour, 2005; Ahn, 2015). In this sense, ANT guided our analytic process by directing attention to the associations that students reported when describing their writing. Conclusions, Expected Outcomes or Findings Preliminary analyses suggest that when mapping their academic writing, students retrospectively unfolded their writing as enacted in between a plethora of actors – both human (e.g., friends, family members, partners, supervisors) and non-human (e.g., GenAI, social media, digital platforms). Moreover, the tentative analysis suggest that GenAI emerges in these reflections as a distinct and recognizable actor, although students’ descriptions are framed predominantly in normative terms; i.e., students seem to orient shared norms on “appropriate” and “inappropriate” use of GenAI, suggesting that the boundaries of acceptable practice are both well-known and actively reproduced in their talk. When discussing misuse, students tend to attribute such practices to the general Swedish pronoun “man” (one), a distancing strategy that reinforces normative compliance. The pattern illustrates the methodological difficulty of capturing empirical data on forms of writing that we conceptualize as “risky”. Some students emphasized that delegating writing tasks to GenAI may hinder learning, framing over‑reliance on AI as a pedagogically counterproductive actor rather than ethically problematic in other ways. One prominent function students described is using GenAI for idea development, as an actor for “bouncing ideas”. In these discussions, GenAI is described as informal, low‑stakes, and easy to engage with, and as offering a “risk‑free” space in the sense that it removes the fear of being negatively judged. These interactions could be contrasted with such that involve teachers, which rather are experienced as slower and more face‑threatening. In the paper presentation, we present the analysis in more detail and discuss implications for academic writing in higher education. References Ahn, S. (2015). Aktör-nätverksteori [Actor network theory]. In A. Fejes, & R. Thornberg, (Eds.), kvalitativ analys [Handbook in qualitative analysis] (2nd ed., pp. 115–130). Studentlitteratur. Bodén, L., & Gunnarsson, K. (2021). Nothing, anything, and everything: Conversations on postqualitative methodology. Qualitative Inquiry, 27(2), 192-197. Barad, K. (2007). Meeting the universe halfway: Quantum physics and the entanglement of matter and meaning. Duke University Press. Barrett, A., & Pack, A. (2023). Not quite eye to A.I.: Student and teacher perspectives on the use of generative artificial intelligence in the writing process. International Journal of Educational Technology in Higher Education, 20(1), 59. https://doi.org/10.1186/s41239-023-00427-0 Brandt, D., & Clinton, K. (2002). Limits of the Local: Expanding Perspectives on Literacy as a Social Practice. Journal of Literacy Research, 34(3), 337–356. https://doi.org/10.1207/s15548430jlr3403_4 Digital Education Council. (2024). Digital Education Council Global AI Student Survey 2024. Digital Education Council https://www.digitaleducationcouncil.com/post/digital-education-council-global-ai-student-survey-2024 Freeman, J. (2025). Student Generative AI Survey 2025 (Policy Note 61). Higher Education Policy Institute (HEPI). https://www.hepi.ac.uk/reports/student-generative-ai-survey-2025/ Kara, H. (2015). Creative research methods in the social sciences. A practical guide. Policy Press. Kim, J., Lee, S.-S., Detrick, R., Wang, J., & Li, N. (2025). Students-Generative AI interaction patterns and its impact on academic writing. Journal of Computing in Higher Education. https://doi.org/10.1007/s12528-025-09444-6 La Scala, J., Sahli, S., & Gillet, D. (2025, April). Stimulating Brainstorming Activities with Generative AI in Higher Education. In 2025 IEEE Global Engineering Education Conference (EDUCON) (pp. 1-10). IEEE. Latour, B. (2005). Reassembling the Social: An Introduction to Actor-Network-Theory. Oxford University Press. Malmström, H., Stöhr, C., & Ou, A. W. (2023). Chatbots and other AI for learning: A survey of use and views among university students in Sweden (Chalmers Studies in Communication and Learning in Higher Education 2023:1). Chalmers University of Technology. https://doi.org/10.17196/cls.csclhe/2023/01 Tsufim, F., & Pomerleau, L. (2024). 6. More is Less?: Using Generative AI for Idea Generation and Diversification in Early Writing Processes. 31. LEd – Network on Language and Education
Paper A Qualitative Study on Business English Undergraduates’ Perceptions of AI-Empowered Learning in Chinese Higher Education Xi'an International Studies University, China, People's Republic of Presenting Author:Abstract Artificial intelligence (AI) is increasingly integrated into higher education and presents transformative opportunities and complex challenges, reshaping how students majored in foreign language programs engage with disciplinary learning (Cao & Phongsatha, 2025), particularly in professionally oriented language programs such as Business English in China. Existing studies often focus on learning outcomes in language communication skills (Wei, Zhang, Ding, Wang, Zhang, Subramaniam, & Meng, 2024; Rintaningrum, 2023), less attention has been paid to how language learners themselves cognitively perceive the meaning, value, and boundaries of AI-empowered learning in Business English Learning (Fuchs, 2022). This qualitative study explores how Business English undergraduates in China perceive and construct the meaning of AI-empowered language learning within their academic and professional development. Moving beyond a general assessment of technology’s utility, the research employs a grounded theory methodology to investigate students’ meaning-making processes regarding AI use in Business English education that characterize student learning experiences. The research conducts an in-depth semi-structured interviews with 24 Business English undergraduates from a Chinese foreign studies university, selected via purposive sampling to ensure diversity in academic performance and AI tools engagement (e.g. generative AI writing assistants, automated feedback systems, and intelligent learning platforms). Through constant comparative analysis by using NVivo 15, the study systematically codes and analyzes the interview transcripts to develop a substantive theory grounded in the data itself. The study identifies three major conceptual categories that reveal a multidimensional tension structure in students’ perception of AI-empowered Business English learning. First, students predominantly construct AI as an instrumental accelerator situated within the fundamental tension between instrumental rationality and humanistic sensibilities. AI is widely perceived as enhancing efficiency, accuracy, and task completion speed, particularly in writing, vocabulary acquisition, listening and speaking, and intercultural business communication practice. However, students simultaneously show concern that over-instrumentalization may weaken deeper learning, critical thinking, and personal expression—elements they regard as central to language education. Second, the findings demonstrate a significant gender-based differences in how AI is symbolically framed in learning practices. Female participants tend to conceptualize AI as a standardizing mentor that provides structured guidance, corrective feedback, and ensures linguistic accuracy, while male participants more often view AI as a creative collaborator that emphasizes its function in supporting ideas and generating content. Importantly, these differences are found to be rooted not in technical competence difference, but in deeply embedded social conditioning regarding gender roles toward technology use. Finally, students experience a distinct crisis of agency though generally positive about AI’s learning potential. Many participants generate anxiety regarding over-dependency and the erosion of learner autonomy. This perceived threat motivates students to actively negotiate boundaries between human creation and machine facilitation. This sense of crisis acts as a catalyst to drive learners to actively negotiate and define the parameters of human-AI collaboration to reclaim their subjective autonomy in the learning process, indicating a critical engagement with AI rather than passive adoption. This study contributes a critical theoretical lens for understanding the complexities of AI integration in discipline-specific language education. It concludes that technological empowerment cannot be a one-dimensional process but must dynamically align with humanistic values of the discipline and proactively support the development of learners’ critical agency (Selwyn, 2022). The findings suggest that effective integration of AI in Business English education requires pedagogical approaches that balance technological efficiency with disciplinary values and support students’ development as autonomous, reflective language users. Methodology, Methods, Research Instruments or Sources Used Research Methods This study adopts a qualitative research design guided by the constructivist grounded theory methodology as developed by Charmaz (2014), aiming to capture how learners construct meaning around AI-empowered learning in the Chinese context. This approach is particularly suited for exploring complex, subjective perceptions and the processes of sense-making, allowing theory to emerge directly from the empirical data. Participants and Sampling: Participants consisted of 24 undergraduate students majoring in Business English at a key Chinese international studies university. A purposive sampling strategy was employed to select information-rich participants that could provide diverse insights into AI-empowered learning experiences. Criteria included varying levels of academic achievement (high, medium, low based on GPA) and self-reported prior experience using AI tools for English learning purposes, including writing assistance, grammar correction, vocabulary learning, and business communication practice. The sample included students from different academic years and balanced gender representation to ensure diverse perspectives. Data Collection: Data were collected through in-depth, semi-structured interviews lasting approximately 45–70 minutes. An episodic interview protocol (Flick, U. (2000). was developed with open-ended questions aiming to elicit detailed narratives about students’ experiences, feelings, attitudes, and conceptual understandings of using AI for Business English learning. All interviews were conducted in Chinese (the participants’ native language), audio-recorded with consent, and later transcribed word by word including non-verbal expressions during the interview to ensure accuracy. Data Analysis: Data analysis followed the constant comparative method characteristic of grounded theory (Corbin & Strauss, 2015). The transcribed interviews were imported into NVivo 15 software for data management and analysis. NVivo 15 software was used to support systematic coding and comparison across cases. Open coding was first conducted to identify significant concepts in the data. Through constant comparative analysis, these initial codes were continually compared within and across transcripts to group them into focused codes and then into higher-level conceptual categories. They were then grouped into higher-level categories through axial coding, and core categories were refined through selective coding. Analytical memos were written throughout the process to enhance reflexivity and theoretical sensitivity. This iterative process of data collection, coding, and comparison continued until theoretical saturation was reached, generating the three core categories that form the study’s findings. Conclusions, Expected Outcomes or Findings Conclusions This study demonstrates that Business English undergraduates’ perceptions of AI-empowered learning are neither uniformly enthusiastic nor uncritically accepting. From an educational research perspective, these findings underscore the importance of focusing on learner agency in discussions of AI-mediated language education. AI tools do not simply “enhance engagement” in a simple way; rather, engagement is shaped by how learners interpret AI’s role in relation to disciplinary values and personal identity as language learners. The identification of AI as “instrumental accelerator,” “standardizing mentor,” and “creative collaborator” offers nuanced categories for analyzing learner positioning. The found gender-related meaning frameworks further suggest that AI adoption with broader sociocultural factors, calling for more effective pedagogical adjustment. For language educators and curriculum designers, the study suggests that AI should be positioned as a pedagogically mediated resource rather than a neutral technological solution. Specific guidance on ethical use, authorship, and reflective AI engagement may help ease learners’ anxiety and support sustainable learning practices. Curriculum design must create opportunities for students to consciously define human-AI collaboration boundaries, thereby transforming anxiety into empowered efficient engagement. Ultimately, the study emphasizes that the true “empowerment” in AI-empowered Business English education lies not in the technology itself, but in designing learning ecosystems that strengthen students’ critical autonomy, discipline-specific humanistic values, and capacity for reflective partnership with intelligent tools. Future research should explore pedagogical interventions designed to mitigate these tensions and actively foster learner agency. To stimulate discussion, this paper raises the following questions for the language learning community: • How can language education reconcile AI-driven efficiency with the cultivation of humanistic and critical dimensions of learning? • In what ways should learner agency be reconceptualized in AI-mediated educational environments? References References: Charmaz, K. (2014). Constructing grounded theory (2nd ed.). Sage Publications. Corbin, J., & Strauss, A. (2015). Basics of qualitative research (4th ed.). Sage. Flick, U. (2000). Episodic interviewing. Qualitative researching with text, image and sound, 75-92. Fuchs, C. (2022). Digital humanism: A philosophy for 21st century digital society. Emerald Group Publishing Limited. Wei, Y., Zhang, Y., Ding, H., Wang, J., Zhang, M., Subramaniam, G., & Meng, W. (2024). The Role of AI in General English and Business English: A Systematic Literature Review of Recent Advancements (2021-2024). Asia Pacific Journal of Business, Humanities and Education, 9(2), 142-164. Cao, S., & Phongsatha, S. (2025). An empirical study of the AI-driven platform in blended learning for Business English performance and student engagement. Language Testing in Asia, 15(1), 39. Rintaningrum, R. (2023). Technology integration in English language teaching and learning: Benefits and challenges. Cogent Education, 10(1), 2164690. Selwyn, N. (2022). Education and technology: Key issues and debates (3rd ed.). Bloomsbury Academic. 31. LEd – Network on Language and Education
Paper Generative AI and Language Education Under Pressure: Process, Agency, and Pedagogy in Art and Design Contexts Emily Carr University of Art and Design, Canada Presenting Author:The rapid integration of generative artificial intelligence (GenAI) into writing, translation, and language-support platforms is reshaping language education across post-secondary contexts. Students increasingly encounter AI-mediated language practices not only in formal language classrooms but also through discipline-specific writing tasks, feedback systems, and professional communication tools embedded in everyday software. Across educational literature, GenAI is widely understood as unavoidable, raising urgent questions about pedagogy, critical judgement, and the conditions under which meaningful learning can occur (Bozkurt et al., 2024; Zewe, 2023). This paper examines the implications of GenAI for language education through the situated context of post-secondary art and design institutions. Art and design programs provide a particularly revealing lens for examining AI-mediated language practices because they foreground process over product, rely heavily on reflective and multimodal writing, and often serve multilingual and internationally mobile student populations. Within these settings, language is not merely a vehicle for communication but a core site of learning, critique, and professional formation. As such, art and design education functions as a high-stakes pressure-test for understanding how GenAI reshapes learner agency, authorship, and linguistic responsibility—concerns that resonate across disciplines and educational systems (Matthews et al., 2023; Park, 2023). The paper is guided by three research questions:
The analysis draws on a multi-year research project based at a Canadian art and design university, including two international environmental scans (Wight et al., 2026). The second scan focused specifically on GenAI impacts on language education within post-secondary art and design institutions, examining policies, pedagogical resources, and professional development materials related to writing, literacy, and AI use. Consistent with findings across creative education and media studies, this research surfaced recurring tensions between policy-driven responses and pedagogical needs, as well as underdeveloped frameworks for supporting educators and tutors tasked with mediating AI use in learning environments (Bender, 2023; Roussel et al., 2024). Conceptually, the paper is grounded in sociocultural theories of language learning, critical literacy, and process-oriented pedagogy. Language education is understood as relational and situated, shaped by power relations, institutional norms, and technological infrastructures rather than as a neutral or purely technical skill. Drawing on scholarship in critical GenAI literacy, GenAI is approached not as a neutral tool but as a socio-technical system that actively participates in shaping norms, values, and exclusions in language use (Elemen, 2024; Hausken, 2024). This perspective is particularly important when considering how AI systems reproduce bias, standardization, and representational harm, including the marginalization of non-standard, racialized, or Indigenous language and cultural practices (Lewis, 2020; Park, 2024). Rather than positioning GenAI as a problem to be managed or a technology to be optimized, this paper argues that art and design writing centres offer transferable pedagogical models for human–AI collaboration in language education. Their emphasis on dialogue, reflection, and care supports learners in developing critical judgement, ethical awareness, and linguistic agency when engaging with AI-mediated language practices. By articulating these practices within a language education framework, the paper contributes insights relevant not only to creative disciplines but also to broader post-secondary language teaching, learning-support, and policy development across Europe and beyond (Andreotti, 2025; Bozkurt et al., 2024). Methodology, Methods, Research Instruments or Sources Used This paper draws on a qualitative, multi-phase research design combining environmental scanning, document analysis, and pedagogical development. The study forms part of an ongoing research project examining the impacts of generative artificial intelligence (GenAI) on language education, writing pedagogy, and learning-support practices within post-secondary art and design institutions. The first phase consisted of an international environmental scan of academic and grey literature published from 2020 onward. Sources included peer-reviewed journal articles, policy documents, institutional guidelines, professional association reports, and public-facing educational resources related to GenAI, writing, literacy, and language education. This scan established a broad landscape of how AI is conceptualized, regulated, and pedagogically engaged across higher education. The second phase narrowed its focus to GenAI impacts on language education within post-secondary art and design institutions and programs. This phase examined materials from institutions in Canada, Europe, and comparable international contexts, including writing and learning centre resources, AI-related curricular guidance, tutor training materials, and disciplinary discussions of authorship, multilingualism, and creative process. This focus allowed for closer analysis of how AI-mediated language practices intersect with process-oriented, multimodal, and often multilingual learning environments. Across both phases, qualitative thematic analysis was used to identify recurring patterns, tensions, and gaps in institutional responses to GenAI. Analytical categories included learner agency, ethical judgement, linguistic diversity, assessment practices, and the balance between policy compliance and pedagogy. Building on these findings, the third phase involves the design and piloting of professional development modules for writing tutors and faculty. These modules emphasize dialogic pedagogy, reflective documentation of AI use, and critical engagement with language norms, bias, and authorship. Iterative feedback from facilitators and participants informed the development of the pedagogical framework presented in this paper. Although empirically grounded in art and design education, the methodology is designed to generate transferable insights for language education across disciplines, particularly in contexts where AI tools increasingly mediate writing, translation, and academic communication. Conclusions, Expected Outcomes or Findings This paper is expected to demonstrate that post-secondary art and design contexts function as critical sites for examining the implications of generative AI for language education more broadly. The findings suggest that while GenAI tools are often framed institutionally as productivity-enhancing or integrity-threatening, they more fundamentally reshape how learners engage with language, authorship, and judgement. One key outcome is the identification of recurring gaps in current institutional responses to GenAI, including limited preparation for tutors and educators, an overreliance on policy-based approaches, and insufficient attention to linguistic diversity, multilingual learners, and process-oriented language development. These gaps risk positioning students as passive users of AI-generated language rather than as active participants in meaning-making. In response, the paper advances a process-oriented framework for tutor education that foregrounds dialogue, metacognition, and ethical reflection. This framework positions GenAI as an object of inquiry within language education rather than a shortcut or substitute for learning. It emphasizes practices such as documenting AI use, articulating linguistic choices, and critically examining bias, translation norms, and authorship. More broadly, the paper argues that writing centres and similar language-support sites offer transferable pedagogical models for human–AI collaboration. Their emphasis on care, learner agency, and reflective practice supports equitable and linguistically just engagement with AI-mediated language practices. These findings contribute to European and international debates on how AI can be integrated into language education in ways that sustain pedagogical integrity, learner autonomy, and social responsibility. References Andreotti, V. (Aiden Cinnamon Tea & Dorothy Ladybugboss). (2025). Burnout from humans. Creative Commons License BY-NC-ND 4.0. Bender, S. M. (2023). Coexistence and creativity: Screen media education in the age of artificial intelligence content generators. Media Practice & Education, 24(4), 351–366. Bozkurt, A., Xiao, J., Farrow, R., Bai, J. Y. H., Nerantzi, C., Moore, S., Dron, J., Stracke, C. M., Singh, L., Crompton, H., Koutropoulos, A., Terentev, E., Pazurek, A., Nichols, M., Sidorkin, A. M., Costello, E., Watson, S., Mulligan, D., Honeychurch, S., & Asino, T. I. (2024). The manifesto for teaching and learning in a time of generative AI: A critical collective stance to better navigate the future. Open Praxis, 16(4). Elemen, J. (2024). Teaching CRITICAL GenAI literacy: Empowering students for a digital democracy. Literacy Today, 42(2), 60–61. Hausken, L. (2024). Photorealism versus photography: AI-generated depiction in the age of visual disinformation. Journal of Aesthetics & Culture, 16(1), 1–13. Lewis, J. E. (2020). Indigenous protocol and artificial intelligence workshops: Position paper 1. Indigenous Protocol and Artificial Intelligence Working Group, Honolulu, Hawaiʻi. Matthews, B., Shannon, B., & Roxburgh, M. (2023). Destroy all humans: The dematerialisation of the designer in an age of automation and its impact on graphic design—A literature review. International Journal of Art & Design Education, 42(3). https://doi.org/10.1111/jade.12460 Park, Y. (2023). Creative and critical entanglements with AI in art education. Studies in Art Education, 64(4), 406–425. Park, Y. S. (2024). White default: Examining racialized biases behind AI-generated images. Art Education, 77(4), 36–45. Roussel, R., Özer, S., Jacoby, S., & Asadipour, A. (2024). Responsible AI in art and design higher education. Wight, K., Burns, L., & Portelance, M. (2026, January 12). Working towards ethical engagement of GenAI in higher education: Insights and recommendations for post-secondary educators. BCcampus Digital Pedagogy Toolbox. https://bccampus.ca/2026/01/12/working-towards-ethical-engagement-of-genai-in-higher-education-insights-and-recommendations-for-post-secondary-educators/ Zewe, A. (2023, November 9). Explained: Generative AI. MIT News. | ||