Â鶹ÒùÔº

The rapid integration of artificial intelligence (AI) into educational practice has introduced a new layer to longstanding debates about decolonising the curriculum. Across schools and universities, AI systems now shape how students search for information, generate ideas and engage with knowledge. This development coincides with ongoing efforts to address curricula that remain structured by colonial histories and Eurocentric epistemologies. The question is not simply whether AI will transform education, but whether it will advance decolonial aims or inadvertently reinforce existing hierarchies.

The movement to decolonise the curriculum has emphasised that knowledge is never neutral. Educational standards, assessment frameworks and research methodologies have historically privileged certain traditions while marginalising others. While seminal works such as the Â鶹ÒùÔº Guide to Decolonising the Curriculum (Moncrieffe et al., 2024) reflect on practices in education across the life course (early years through to higher education), little attention is paid to the profound influence that AI can have on the decolonising agenda.

AI as epistemic reproduction

Large language models and other AI tools are trained on vast datasets that reflect existing patterns of publication, citation and digital archiving. These patterns are unevenly distributed globally, with scholarship from the Global North more likely to be digitised, indexed and widely cited. Consequently, AI systems may reproduce dominant epistemic frameworks even when they appear to offer expansive access to information. For instance, research by Arora et al. (2023) demonstrates how algorithmic systems can encode and amplify racialised and gendered biases. When such systems are embedded in educational contexts, they risk presenting historically dominant perspectives as universal knowledge.

‘Algorithmic systems can encode and amplify racialised and gendered biases. When such systems are embedded in educational contexts, they risk presenting historically dominant perspectives as universal knowledge.’

Between recolonisation and epistemic pluralism

Furthermore, some scholars have argued that contemporary AI systems are also embedded within ecological forms of coloniality through its environmental footprint and extractive infrastructures, though AI is not inherently recolonising in pedagogical terms. Its effects are contingent on how it is designed, governed and used. AI tools can assist educators in surfacing scholarship from underrepresented regions, translating texts across languages and enabling students to engage with multiple intellectual traditions, potentially supporting a more pluralistic curriculum. However, contemporary forms of data extraction may constitute a new phase of coloniality (Zembylas, 2021). Their concept of Digital-Territorial Coloniality highlights how global inequalities shape whose knowledge is captured and whose labour underpins digital infrastructures. In educational settings, this raises important questions about who contributes to AI training data and who benefits from its outputs.

For the educational community, the duality of AI therefore demands careful scrutiny. It is tempting to frame AI as either a transformative tool for inclusion or a threat to epistemic justice; but in practice, it functions as a contested space in which both possibilities coexist. Decolonising the curriculum in the age of AI requires a critical approach to data sources, algorithmic design and classroom implementation. Educators must consider how AI tools shape what students encounter as authoritative knowledge and how they might be used to foreground marginalised perspectives rather than obscure them (Allison, 2025).

Pedagogical responsibility in the age of AI

There is also a need for robust AI literacy within teacher education and professional development, as understanding how AI systems are trained, what biases they contain and how they can be critically evaluated is now central to . Without such literacy, there is a risk that AI will be adopted uncritically, reinforcing rather than challenging dominant narratives. Educators and institutions might begin by asking: whose epistemologies are most visible in the AI systems we adopt? What environmental and labour conditions underpin their operation? How are students being supported to treat AI outputs as situated and contestable rather than neutral? Framed in this way, AI literacy becomes inseparable from questions of power, representation and material infrastructure.

The question, then, is not whether AI will decolonise or recolonise education, but how educational communities will shape its role. For researchers, policymakers and practitioners, this moment presents an opportunity to embed decolonial principles into emerging digital infrastructures. AI should be understood not as a neutral technological development but as part of an ongoing struggle over knowledge, authority and representation. Engaging critically with this duality will be essential if AI is to contribute to more equitable and pluralistic educational futures.

References

Allison, J. (2025). Addressing the false neutrality of computing in computing education. In A. A. Shafi, A. Twyman-Ghoshal, A. C. Afriyie, S. R. Copland, & O. El Masri (Eds.), Innovations in decolonising the curriculum: Multidisciplinary perspectives. Emerald Group Publishing.

Arora, A., Barrett, M., Lee, E., Oborn, E., & Prince, K. (2023). Risk and the future of AI: Algorithmic bias, data colonialism, and marginalization. Information and Organization, 33(3).

Moncrieffe, M. L., Fakunle, O., Kustatscher, M., & Rost, A. O. (Eds.). (2024). The Â鶹ÒùÔº guide to decolonising the curriculum: Equity and inclusion in educational research and practice. Emerald Group Publishing. /publication/the-bera-guide-to-decolonising-the-curriculum

Zembylas, M. (2021). A decolonial approach to AI in higher education teaching and learning: Strategies for undoing the ethics of digital neocolonialism. Learning, Media and Technology, 48(1), 25–37.