BLESSING MAREGERE, SENIOR WORK-BASED LEARNING TUTOR AND PROGRAMME LEADER, LEEDS TRINITY UNIVERSITY, UK
Introduction: Creativity at a crossroads
As generative artificial intelligence (GenAI) impacts education, the question confronting educators is not whether GenAI should be in the classroom but how it should coexist with human creativity. GenAI-powered writing assistants, adaptive learning platforms and automated feedback systems offer efficiency and personalisation; however, they also risk limiting the human aspects of teaching – curiosity, empathy and dialogue. For many educators, an emerging paradox exists: while GenAI can support learning and creativity, it may also weaken learners’ skills and promote standardised thinking (Zhai et al., 2024).
This perspective piece argues that dialogic pedagogy – learning through structured, purposeful dialogue – provides a strong, human-centred counterbalance. Rooted in traditions of reflective inquiry and shared construction of meaning, the dialogic pedagogy approach fosters the very skills that automation often overlooks: critical questioning, imagination and ethical awareness (Alexander, 2020; Mercer et al., 2019). Drawing on research into professional discussion, a form of dialogic assessment used in apprenticeships, the article demonstrates how structured dialogue boosts teacher and learner creativity and offers practical strategies for incorporating GenAI without weakening human agency.
The promise and challenges of GenAI in education
GenAI has become a transformative tool in teaching, learning and assessment. From intelligent tutoring systems to AI-assisted assessment, the technology offers innovative opportunities for data-driven insights and personalised feedback (Chiu et al., 2024). However, the same capabilities raise ethical and pedagogical concerns when algorithms mediate what and how learners learn, with creativity at risk of becoming instrumental – reduced to patterns that machines can recognise and reproduce (Acar, 2023).
Recent empirical studies highlight the tension between augmentation and automation. Zhou and Peng (2025) found that while AI tools can stimulate learners’ creative engagement, overreliance may weaken intrinsic motivation and self-reflection, especially when AI literacy is low. Similarly, Zhai et al. (2024) concluded that sustained use of conversational AI systems can limit critical thinking by discouraging learners from generating independent explanations. These findings echo long-standing pedagogical warnings about passive learning environments that prioritise correct answers over critical processes (Scott et al., 2006).
At its core, the challenge is epistemic: how can educators ensure that AI enhances rather than replaces the dialogic, relational and uncertain aspects of learning that characterise creativity itself?
Dialogic pedagogy: Reclaiming human connection
Dialogic pedagogy offers a conceptual framework for addressing this challenge. Defined by Alexander (2020) as teaching that harnesses the power of talk to stimulate and expand students’ thinking and promote their learning, it emphasises mutual respect, open questioning and the co-creation of knowledge, understanding and feedback. Mercer et al. (2019) further describe dialogue as a shared space where learners think together, regulating both cognitive and socio-emotional aspects of learning.
Dialogic environments differ from traditional ‘delivery’ models because they present knowledge as provisional, negotiated and relational. Uçan et al. (2023) emphasise that dialogic learning fosters emotional safety and metacognitive development by inviting learners to express uncertainty and consider alternative perspectives. In this way, dialogue sustains creativity not as spontaneous inspiration but as socially constructed innovation.
The professional discussion as a dialogic practice
In HE (higher education) and apprenticeship education, the professional discussion exemplifies dialogic pedagogy and assessment in practice. Used extensively in apprenticeships and professional programmes, professional discussions consist of structured yet conversational exchanges between learners and their assessors to evidence knowledge, skills and behaviours. My research into professional discussions has demonstrated that, when conducted dialogically, these discussions serve as a powerful space for creativity, reflection and identity formation (Maregere, 2026).
Unlike written assignments, which can be easily copied or generated by AI, dialogic assessments focus on authentic dialogue, requiring learners to justify decisions, connect theory to practice and reflect in real time. This immediacy encourages creativity by compelling learners to rehearse professional judgement, think aloud, negotiate meaning and adapt responses dynamically. It also fosters tutors’ creativity, as they co-construct understanding through adaptive questioning and ethical engagement listening.
Such dialogic assessments align with Chinn’s (2017) framework for argumentation and explanation, where learning is seen as a process of generating, evaluating and refining ideas through social interaction. In dialogic assessments, knowledge is not simply transmitted but performed – enacted through dialogue that values curiosity and perspective-taking (Maregere, 2026).
Creativity as a dialogic and ethical process
Creativity in education is often viewed as producing novel ideas or artefacts. However, dialogic pedagogy situates creativity within an ethical and relational process. Alexander (2020) reminds us that dialogue involves both listening and speaking, encouraging learners to recognise others’ voices as legitimate sources of meaning. This ethical stance becomes increasingly vital in the AI era, where synthetic voices simulate but do not replicate human intentionality.
Acar (2023) argues that assessing creativity in the age of AI must include human judgement, originality and moral reasoning – elements that are hard to automate. Dialogic pedagogy supports these by encouraging learners to justify their creative choices and critique the ethical implications of AI use. In classroom practice, this might involve asking: Whose perspective is amplified or silenced when AI produces content? How does relying on AI influence my authorship?
By emphasising reflection and empathy, dialogic teaching redefines creativity as a process of collective meaning-making rather than simply individual output. This view is consistent with sociocultural theories of learning that see dialogue as the link between internal thought and shared understanding (Kim and Wilkinson, 2019).
When AI dominates: The risk to authentic voice
As generative AI becomes more embedded in education, assessment and feedback, risk becomes less relational and more standardised. Although GenAI can scale personalised support, timely feedback and new forms of assessment, these benefits also come with concerns about depersonalisation, over-standardisation and the weakening of human agency in learning (Yan et al., 2024; UNESCO, 2023). Creative learning depends partly on productive struggle and learning through failure, both of which help learners to refine ideas, test judgment, and develop originality (Manalo and Kapur, 2018; Henriksen et al., 2021). In a systematic review, Zhai et al. (2024) found that overreliance on AI dialogue systems can undermine students’ decision-making, critical thinking and analytical reasoning. Likewise, Nelson et al. (2025) found that students were more concerned that GenAI might hinder the development of their own writing skills than about being caught using it dishonestly.
The risk to authentic voice is therefore not only technical but also pedagogical and cultural. When learners depend heavily on machine-generated phrasing and reasoning, there is a risk that academic work becomes polished yet derivative, weakening students’ sense of authorship and confidence in their own judgement. By contrast, dialogic pedagogy and dialogic assessment validate student voice through shared responsibility, conversation and reflection, rather than treating feedback as something simply delivered to passive recipients (Bain, 2010; Laurillard, 2013; Carless and Boud, 2018). These approaches preserve the imperfections, nuance and developmental movement that are central to authentic learning.
Educators should therefore position AI as a support for learning, and not a substitute for human interpretation. AI can assist with brainstorming, idea generation, language support, data handling and low-stakes formative feedback, but the interpretive, relational and ethical dimensions of learning should remain human-led (UNESCO, 2023; Yan et al., 2024). In practice, this means using AI to extend learners’ thinking while ensuring that judgement, meaning-making and the formation of academic voice remain rooted in human dialogue and professional acumen.
Integrating AI dialogically: Practical strategies
For educators in classrooms and workplaces seeking to incorporate AI into education while protecting creativity, dialogic pedagogy principles offer practical guidance. The following strategies combine insights from current research and professional practice.
1. Make AI a topic for dialogue and not a substitute for it
Encourage learners to share their experiences with AI, including its benefits, limitations and ethical concerns. Structured class debates or reflective journals can help students to develop AI literacy (Zhou and Peng, 2025) and foster a critical perspective towards the technology instead of passively accepting it.
2. Design dialogic assessments resistant to automation
Assessments that require explanation, justification or improvisation – such as interactive orals, professional discussions or viva voces – are inherently resilient to AI imitation. They also encourage creativity by recognising the process, and not just polished products (Scott et al., 2006).
3. Use AI to extend and not replace human feedback
AI-driven tools can assist in recognising trends in student writing or providing low-stakes formative feedback. However, educators should present these insights as prompts for discussion within subsequent human feedback sessions, aligning with Mercer et al.’s (2019) concept of shared regulation of learning.
4. Cultivate metacognition and ethical reflection
Encourage learners to articulate how and why they used AI in their work. Reflective prompts, such as ‘What decisions did you make that AI could not?’ or ‘How did dialogue with others influence your thinking?’, reinforce ownership and accountability (Garcia Ramos, 2025; QAA, 2025; Ajjawi and Boud, 2017; Carless and Boud, 2018).
5. Model dialogic teaching with transparency and humility
Teachers can demonstrate human-centred innovation by sharing their own uncertainties about AI, showing curiosity rather than authority. As Uçan et al. (2023) observe, such vulnerability enhances socio-emotional trust and invites learners into co-inquiry.
From automation to co-creation: Rethinking innovation
Educational innovation is often linked to technological adoption. However, true innovation is pedagogical rather than mechanical. It arises from relationships that boost human capacity for meaning, creativity and care. Dialogic pedagogy redefines innovation as a collaborative process – where teachers and learners work together to navigate complexity through discussion, reflection and ethical judgement.
This redefinition aligns with global efforts to humanise technology-enhanced learning. Recent scholarship on AI in education emphasises human-centred, critically reflective adoption. Chiu et al (2024) extend AI competency beyond technical knowledge and also include confidence and a self-reflective mindset, while UNESCO (2024) places human agency and accountability at the centre of Teacher AI competency. Similarly, Kim and Wilkinson (2019) highlight that dialogic pedagogy across cultures fosters intercultural empathy and collective problem-solving – precisely the skills needed to thrive in an AI-mediated world. This literature suggests that AI should support not replace, educator professional judgement and the relational element of teaching and learning (UNESCO, 2023; OECD, 2026).
By integrating AI into dialogic frameworks, education can move beyond simple debates of ‘AI versus teacher’ to foster collaborative human–machine partnerships that appreciate creativity, both individually and collectively.
Implications for practice
For classroom teachers, trainers and curriculum leaders, reclaiming creativity in the AI era requires intentional design choices that prioritise dialogue at the centre of learning. The following implications may guide practice:
- Curriculum design: Embed structured dialogue, such as Socratic seminars, interactive orals, selfie videos or peer learning circles, where learners articulate reasoning and critique AI-generated material
- Assessment: Prioritise tasks that assess process rather than product – reflective commentaries, collaborative projects and oral defences that demonstrate learners’ developing understanding
- Professional learning: Foster communities of practice where educators experiment with dialogic assessment, exchange examples and critically evaluate AI tools collaboratively
- Policy and leadership: Institutional policies should prioritise dialogic and creative assessment as indicators of quality, resisting over-standardisation that values efficiency over depth.
These shifts require time and trust but are strongly connected with the idea of linking research and practice for the benefit of all learners.
Conclusion: Reclaiming the human in the machine age
Creativity is not a luxury but a necessity in an era dominated by automation. As AI becomes more integrated into education, teachers must reclaim creativity as a dialogic and ethical act – one that values human unpredictability, empathy and co-creation.
Dialogic pedagogy provides both a philosophy and a practical approach for this reform. By engaging learners in meaningful conversations, encouraging reflection and revisiting the appeal of algorithmic certainty, educators can ensure that technology enhances rather than reduces humanity.
In this sense, dialogic teaching is not a retrospective return to pre-digital methods but a forward-looking innovation – a human-centred reimagining of what it means to learn, teach and create in the AI era.
The examples of AI use and specific tools in this article are for context only. They do not imply endorsement or recommendation of any particular tool or approach by the Department for Education or the Chartered College of Teaching and any views stated are those of the individual. Any use of AI also needs to be carefully planned, and what is appropriate in one setting may not be elsewhere. You should always follow the DfE’s Generative AI In Education policy position and product safety standards in addition to aligning any AI use with the DfE’s latest Keeping Children Safe in Education guidance. You can also find teacher and leader toolkits on gov.uk.
