Conference Agenda

The increasing presence of artificial intelligence (AI) in education has generated new questions for teacher education and professional training across Europe. Although AI is frequently presented as a transformative development for teaching and learning, uncertainty persists regarding how teacher education prepares teachers to engage with its pedagogical, ethical, and professional implications. A substantial part of research on AI in education continues to prioritise technical capabilities or tool development. Engagement with educational theory and teachers’ professional learning needs, therefore, often remains limited (Zawacki-Richter et al., 2019; Chen et al., 2020; Tang et al., 2023). In language education in particular, recent AI applications (e.g., automated feedback, text generation and analytics) raise questions about assessment, authorship, feedback practices and teachers’ pedagogical decision-making.

Several literature reviews have highlighted that AI-in-education research is largely shaped by STEM-oriented perspectives and frequently lacks pedagogical grounding, resulting in limited relevance to everyday teaching practice (Kabudi et al., 2021; Bates et al., 2020). As a consequence, teachers are increasingly expected to respond to AI-related change without clear guidance on how these developments relate to professional judgement, assessment practices, and ethical responsibility.

More recent work foregrounding teachers’ perspectives (Tan, 2024; Kashif et al., 2025) suggests that challenges related to AI are not primarily technical, but professional: how teachers exercise pedagogical judgement, justify assessment decisions, and navigate responsibility when AI systems shape feedback and evaluation practices (Tan, 2024). These concerns align with emerging international competency framings that foreground human agency, ethics, and professional learning as core dimensions of teachers’ AI preparedness (UNESCO, 2024).

This paper examines how AI is addressed within teacher education and professional training contexts and explores the professional learning challenges this creates for teacher preparation. The guiding research question is: How is AI being addressed within teacher education and professional training, and what professional learning challenges does this create for preparing teachers for practice?

Sub-questions specify (1) how AI is framed in these learning contexts (e.g., tool uptake vs. professional judgement/ethics), and (2) which challenges participants describe for assessment, feedback, and accountability?

The study is grounded in contemporary theories of teacher professional learning, which conceptualise learning as an ongoing, situated, and reflective process. From this perspective, teacher education is understood as a key site for developing teachers’ professional judgement, pedagogical reasoning, and ethical awareness. Specifically, we draw on scholarship that treats professional learning as situated and practice-linked (Borko, 2004) and shaped by complex interacting systems (Opfer & Pedder, 2011), emphasising collaborative reflection and inquiry as conditions for meaningful change. AI is approached as a pedagogical and professional issue that reshapes teaching practices, assessment processes, and professional responsibility, rather than as a question of technical competence or tool adoption.

We therefore treat “AI-supported teaching” as a boundary problem of professional practice: teachers must interpret AI outputs, decide what counts as valid evidence, and remain accountable for consequential judgements—especially in assessment.

Empirically, the paper draws on data collected in in-service teacher education and professional training contexts. AI is examined in relation to reflection, assessment-related judgement, and collaboration. The research focuses on how AI is framed and discussed within teacher education and how this framing shapes teachers’ professional learning experiences. By situating the analysis within a European context, the paper contributes to ongoing discussions on how educational research and teacher education respond to changing conditions of teaching and learning, foregrounding professionalism, ethical reflection, and pedagogical judgement. The contribution is an empirically grounded account of how AI is positioned in professional learning and a set of implications for designing teacher education that prioritises judgement, assessment literacy, and accountability rather than stand-alone tool training.

There are high levels of linguistic diversity in Europe, with national, foreign, border, regional minority and migrant languages growingly co-existing in classrooms. It is therefore important for teachers to be able to manage the multilingual practices of their pupils in order to effectively engage with multilingualism in the classroom (Arocena et al., 2015).

In the north of the Netherlands, where this study is based, Dutch is the official language, English has a significant presence as a foreign and prestige language, German is spoken across the border and taught in school, Frisian and Low Saxon are spoken as minority languages, and many other home languages are spoken. While the role of teacher training is to prepare future teachers for the realities of our society, pre-service teachers (PSTs) have reported feeling underprepared to engage with multilingual practices in education and do not receive enough training to allow them to positively use multilingualism in their classrooms, including in the Netherlands (Henehan & Duarte, 2024; Llompart et al., 2023; Robinson-Jones et al., 2022).

In this study we address PSTs’ engagement with multilingualism using the dominant language constellation (DLC) framework. This is an approach to multilingualism which focuses on the core languages which are used to fulfill an individual’s, community’s or nation’s communicative needs (Aronin, 2006; 2019). As such, this framework is useful to map multilingualism in education, as the entire linguistic repertoire of a class group may be too large for a PST to manage effectively, but by using a DLC framework, the PST will be able to visualise all the languages used by their pupils and focus on the core languages, without needing to be capable of using all of the languages spoken by their pupils. Furthermore, DLCs can be created as artworks mapping individuals’ language identities, allowing participants to not only describe their language use, but also the connection to their personal identities, context of language use and language cognitions (Ibrahim, 2022; Prada, 2024).

Artificial intelligence (AI) can be employed in connection with multilingualism and DLCs. AI is a useful tool which has not yet been thoroughly explored in the field of multilingualism, but which holds strong potential for the development and inclusion of multilingual pedagogies in education (Holmes & Tuomi, 2022; Luckin, 2018). Advancing knowledge in this topic is relevant because the use of digital tools including AI may support PSTs in engaging with multilingual education (Holmes & Tuomi, 2022; Luckin, 2018; Smeins et al., 2022). AI is particularly promising for PSTs as it can be tailored to the specific context of their classrooms. Furthermore, as they are training to become teachers during the current AI boom, the current study also develops their critical AI literacy at a relevant moment and deepens our understanding of the uses of AI for education (Péres-Paredes et al., 2025).

The goal of this study is to engage PSTs with the multilingualism present in their classrooms and with their pupils’ multilingual identities using the DLC framework and AI. We aim to answer the following research question:

How can AI be used to help PSTs engage with multilingual educational practices for their specific contexts by using the DLCs created by their pupils?

Since the 2022 public release of ChatGPT, educational research has expanded at a remarkable speed, accompanied by a proliferation of seductive conceptual labels for learning, such as AI literacy and especially critical AI literacy. In the Nordic context specifically, recent scholarship highlights how AI’s rise has pushed AI literacy to the center of debate, with the agenda now dominated by questions of which competencies learners need to navigate a data-driven society (Velander et al., 2024). While this terminological surge suggests heightened attention to AI, it also raises a central question for education research in general and literacy studies in particular: what kinds of AI-related studies are being prioritized, which concerns are driving the research agenda, and what forms of educational action do these priorities enable, foreclose, or normalize?

Decades ago, Selfe (1999) called on literacy and language scholars to pay critical attention not only to technology but also to the discourses on technology that shape educational policy. This call went beyond questioning whether educators should incorporate technologies in their teaching. Instead, it urged reflection on the material, institutional, and ideological conditions that shape how technologies function in educational contexts. Yet more recently, Decuypere et al. (2021) observed that educational research that adopts a critical gaze towards technologies remains very limited; a constraint also visible in the Nordic tradition, where Burnett (2010) noted the dominance of narrower, instrumental conceptions of digital technology and literacy. However, without sociotechnical grounding, critical AI literacy risks becoming a thin proceduralism – lists of tips for prompt-writing and source-checking – thus missing the opportunity to re-learn important lessons about literacy.

This paper offers a critical literature review of research on literacy and artificial intelligence (AI) produced in the Nordic countries since the public launch of generative systems. Building on the previous diagnosis, we ask whether the same criticality gap characterizes Nordic scholarship on AI by examining 1) how studies frame their research problems and purposes, 2) what study designs and methodological approaches they draw on, and 3) how they conceptualize both AI platforms and literacy. Our initial hypothesis is that, although Nordic scholars are certainly paying attention to AI, the critical focus is absent, with most scholarship adopting instrumental framings that regard AI systems as “just a tool” (McKnight and Shipp, 2023), emphasizing applications over in-depth, holistic analysis and leaving core concepts in literacy and language research undertheorized. Second, we build an interdisciplinary bridge by situating these studies within longer-standing debates in literacy studies, platform studies, and critical technology studies to propose directions for what the “critical” in “critical AI literacy” could entail for literacy and language researchers.

This study draws on sociocultural perspectives that conceptualize both literacy and technology as ideological (Street, 1984), which means their acquisition and use are seen as more than “using a tool,” and instead as a complex situated process, entangled with epistemology, power, and identity. We also argue that literacy’s cultural value rests not only on immediate use but on its attachment to “imagined futures” and “the formation of particular kinds of people or societies” (Nichols et al., 2024, p. 212). In this sense, literacy education has long operated as a speculative project, repeatedly tied to promises of moral improvement, economic advancement, and empowerment. Against this backdrop, we argue that meaningful literacy research should not be dedicated to continually redefining what “literate” means in the face of each new technology, but rather to develop theory that can account for how core constructs—such as agency, voice, or authorship—are being reconfigured. It should also question, resist, and test totalizing narratives, whether they appear as enduring literacy myths or as end-of-the-world forecasts.

Against the backdrop of global educational digitalization, EFL writing instruction in junior high schools often faces challenges: students struggle to produce relevant, coherent texts due to passive teacher-led interactions, while AI tools (e.g., ChatGPT) remain underutilized as interactive teaching aids. This study focuses on Grade 9 Chinese EFL learners and addresses three core research questions:

1. How can AI-facilitated open dialogue between teachers and students be designed to target gaps in writing relevance (e.g., topic alignment) and coherence (e.g., logical connections) among Grade 9 EFL learners?
2. What theoretical framework integrates sociocultural theory (Vygotsky, 1978) and AI-enhanced instructional design to guide this dialogue-based writing intervention?
3. What implications does this practice hold for international EFL contexts (especially European multilingual classrooms) grappling with similar writing challenges?

The theoretical framework draws on two pillars:

• Sociocultural Theory: It posits that learning occurs through social interaction (Vygotsky, 1978). The intervention uses AI (e.g., AI-powered dialogue prompts) to expand the "zone of proximal development": teachers and AI co-facilitate open dialogue (e.g., asking "How does this detail support your thesis?"), guiding students to refine writing logic collaboratively.
• AI-Enhanced Language Instruction Framework (He et al., 2023): This framework emphasizes AI’s role in personalizing interaction—AI analyzes students’ draft coherence in real time, generating targeted dialogue prompts (e.g., "Add a transition sentence here to link ideas") to supplement teacher feedback.

The European/international dimension is critical:

• Alignment with European Multilingual Education: European classrooms (e.g., bilingual schools in Finland) face similar writing coherence challenges in L2 instruction. This intervention’s "AI+teacher dialogue" model can be adapted to European contexts, leveraging AI to support multilingual learners’ logical expression across languages.
• Response to Global EFL Trends: The study aligns with the Council of Europe’s (2020) "Common European Framework of Reference for Languages" (CEFR) focus on "communicative coherence," offering a scalable AI-integrated approach for international EFL writing instruction.