Reconstituting Teaching in AI-Mediated Education: Epistemic Authority, Labour, and the Post Digital Profession

 

Abstract

The rapid proliferation of artificial intelligence (AI) in education has generated extensive debate regarding its implications for teaching. Much of this discourse focuses on teacher attitudes, adoption, and classroom practice. However, such approaches are insufficient. Drawing on sociotechnical and post digital perspectives, this study reconceptualises AI as a structural force that reconfigures the epistemic, labour, and institutional conditions of teaching, rather than merely a tool to be adopted. A triple tension framework is developed—efficiency versus epistemic integrity, innovation versus deprofessionalisation, and support versus precarity—to explain how AI reshapes the profession beyond surface-level practice. The concept of teachers as epistemic mediators in AI-rich environments is advanced, repositioning professional authority from knowledge transmission to the validation, interpretation, and orchestration of machine-generated knowledge. The analysis concludes that teaching is not being displaced by AI but is undergoing a contested reconstitution, the outcomes of which depend on governance, institutional design, and critical engagement with educational technologies.

Keywords: artificial intelligence, teaching profession, epistemic authority, post digital education, teacher labour, EdTech

1. Introduction

Artificial intelligence (AI) has transitioned rapidly from speculative discourse to routine educational practice. Generative AI systems are now commonly employed for lesson planning, assessment design, and student learning support. In response, research has primarily focused on teacher attitudes, readiness, and classroom integration. While this body of work is valuable, it risks framing AI as an external innovation that teachers must merely respond to.

A different position is adopted here: AI should be understood as a structural intervention in the organisation of educational work, rather than as a discrete pedagogical tool. The central question is not whether teachers adopt AI, but how AI is reconstituting the conditions under which teaching occurs.

To address this issue, a theoretically grounded account is developed to examine how AI reshapes the teaching profession across three interrelated domains:

• Epistemic authority (what counts as knowledge and who validates it)
• Professional labour (what teachers do and how their work is organised)
• Institutional power (how education is governed and mediated through technology)

This approach moves beyond perception-based accounts to offer a conceptual intervention aligned with post digital and sociotechnical scholarship.

2. From Adoption to Reconstitution: A Sociotechnical Framing

Mainstream research on AI in education tends to follow an adoption logic, examining factors that influence whether and how teachers use AI tools. This includes perceived usefulness, ease of use, and professional readiness. While such frameworks provide useful insights, they are limited in two ways.

First, they assume that technology is external to professional practice, rather than constituting it. Second, they privilege individual agency while underplaying structural conditions such as policy mandates, platform infrastructures, and commercial interests.

A sociotechnical perspective offers an alternative. It conceptualises education as an assemblage of human and non-human actors, in which technologies actively shape practice, knowledge, and power relations. From this perspective, AI is not simply integrated into teaching; it reorganises teaching as practice.

This aligns with post digital education scholarship, which rejects the separation of “digital” and “non-digital” and instead focuses on their entanglement. AI, in this sense, is part of a broader shift toward datafied, automated, and platform-mediated education.

3. Reconfiguring Teacher Labour

AI is frequently positioned as a solution to teacher workload. Tools that automate lesson planning, generate feedback, and streamline administrative tasks are widely promoted as efficiency-enhancing innovations. However, this narrative warrants critical examination.

Rather than reducing labour, AI redistributes it. Teachers are increasingly required to engage in:

• Editing and validating AI-generated materials
• Designing prompts and refining outputs
• Interpreting algorithmically produced data

This reflects a broader shift in digital labour, in which human work is oriented toward supervising and correcting machine processes. Such changes may alter not only the tasks teachers perform, but also how their expertise is valued.

Furthermore, efficiency gains may result in labour intensification. As AI enables faster production of teaching materials, institutional expectations may increase, requiring greater differentiation, responsiveness, and documentation. In this context, AI may simultaneously alleviate and exacerbate workload pressures.

These dynamics are unevenly distributed. In resource-rich environments, AI can serve as a productivity tool, whereas in under-resourced contexts, it can introduce additional burdens due to a lack of infrastructure, training, or support. The impact of AI on teacher labour is therefore context-dependent and stratified, rather than universally beneficial.

4. Epistemic Authority in AI-Mediated Classrooms

A significant transformation concerns the epistemic role of teachers. Traditionally, teachers have been positioned as authoritative sources of knowledge. AI disrupts this arrangement by enabling students to generate content independently and access explanations on demand.

This development does not eliminate the teacher's role but reconfigures it. Teachers are increasingly required to act as epistemic mediators, responsible for:

• Evaluating the accuracy and reliability of AI outputs
• Guiding students in interpreting machine-generated knowledge
• Designing tasks that foreground critical engagement rather than reproduction

This shift reflects a broader transformation in the nature of knowledge. In AI-mediated environments, knowledge is no longer scarce or stable but is abundant, provisional, and algorithmically produced. The primary challenge for education is not access, but validation and meaning-making.

Claims that AI leads to a decline in critical thinking should be approached with caution. Such assertions often rely on perception rather than longitudinal evidence and may reflect anxiety regarding changing epistemic norms. A more productive approach is to recognise that AI is redefining what constitutes thinking, thereby requiring new pedagogical and assessment strategies.

5. Professional Identity: Adaptation, Not Collapse

Public discourse frequently frames AI as triggering a crisis in teacher identity. Although some educators’ express uncertainty or concern, the notion of a universal identity crisis is overstated.

Instead, evidence suggests a process of differentiated adaptation. Teachers respond to AI in varied ways depending on their context, experience, and institutional environment. Some resist or limit its use, others integrate it pragmatically, and a smaller group actively innovates.

Professional identity is therefore not collapsing but being renegotiated. New roles are emerging, including:

• Designers of AI-supported learning environments
• Facilitators of digital and AI literacy
• Interpreters of data-driven insights

These developments indicate an expansion of professional identity, albeit one that requires new forms of expertise and support. The key issue is not whether teachers remain relevant, but how their expertise is redefined in relation to AI systems.

6. The Political Economy of AI in Education

Understanding AI in education requires attention to its political and economic context. AI systems are typically developed and deployed by private companies, often operating within platform-based business models that rely on data extraction and monetisation.

This raises important questions about:

• Data ownership and governance
• The role of commercial actors in shaping education
• The implications of algorithmic decision-making

In this context, teachers are not only users of technology but also contributors to data ecosystems. Their interactions with AI systems generate data that can be used to refine algorithms and develop new products. This positions teachers within a form of data labour, often without clear recognition or compensation.

Moreover, the adoption of AI is frequently driven by policy and market pressures rather than pedagogical need. These dynamics can create tensions between professional judgment and institutional expectations, contributing to a sense of reduced autonomy.

A political economic perspective thus highlights that AI is not neutral but embedded in power relations that shape educational practice.

7. Beyond the Human–AI Binary

A common framing of AI in education is the division of labour between humans and machines, where AI handles routine tasks and teachers focus on relational and higher-order functions. Although intuitively appealing, this binary is overly simplistic.

In practice, the boundaries between human and machine work are increasingly blurred. Teachers use AI creatively, while AI systems increasingly simulate forms of interaction traditionally associated with human teaching. This results in hybrid pedagogical practices that cannot be easily categorised.

Rather than maintaining a strict division, it is more productive to conceptualise teaching as an entangled practice in which human and machine contributions are dynamically negotiated. The key question is not what AI or teachers do separately, but how their interactions reshape educational processes.

8. A Triple Tension Framework

To synthesise the analysis, a triple tension framework is proposed to capture the key dynamics shaping the teaching profession in AI-mediated education:

8.1 Efficiency vs Epistemic Integrity

AI enhances productivity, but challenges established notions of authorship, originality, and learning.

8.2 Innovation vs De-professionalisation

AI enables new forms of teaching while potentially reducing autonomy and redistributing expertise.

8.3 Support vs Precarity

AI provides practical support but introduces new forms of surveillance, dependency, and labour vulnerability.

These tensions are not temporary but are constitutive of the current moment. They reflect the broader transformation of education in a datafied, platform-mediated society.

9. Conclusion

AI is not merely an innovation to be adopted within existing educational frameworks. It is a structural force that is reshaping the epistemic, labour, and institutional foundations of teaching. Teachers are not being replaced, but their roles are being redefined in relation to intelligent systems.

This redefinition is uneven, contested, and ongoing. The outcomes will depend on how AI is governed, how institutions support teachers, and how educators critically engage with technological change.

Future research should move beyond questions of adoption and attitude to examine the structural conditions shaping teaching in AI-mediated environments. This approach is necessary to understand not only how teachers respond to AI, but also how the profession itself is being transformed.

References

Holmes, W., Bialik, M., & Fadel, C. (2022). Artificial intelligence in education. Center for Curriculum Redesign.

Jandrić, P., Knox, J., Besley, T., Ryberg, T., Suoranta, J., & Hayes, S. (2018). Postdigital science and education. Educational Philosophy and Theory, 50(10), 893–899.

Selwyn, N. (2016). Education and technology: Key issues and debates. Bloomsbury.

Selwyn, N. (2023). Education and artificial intelligence. Routledge.

Srnicek, N. (2017). Platform capitalism. Polity.

UNESCO. (2023). Guidance for generative AI in education and research. UNESCO.

Williamson, B. (2017). Big data in education. Sage.

Williamson, B., & Eynon, R. (2020). Historical threads and future directions in AI in education. Learning, Media and Technology, 45(3), 223–235.