Artificial intelligence is transforming development practice, university education and graduate work. Development studies must respond not by chasing technological novelty, but by strengthening its distinctive capacity to interrogate power, inequality, ethics and institutional change.
Artificial intelligence is no longer a distant technological development. It is already influencing how governments allocate public services, financial institutions assess risk, humanitarian organisations identify need, employers organise work and universities evaluate learning.
These developments make AI an urgent concern for development studies. The central question is not simply whether development professionals should learn to use new tools. It is whether students can understand how AI systems redistribute authority, reproduce inequality and shape decisions about people and communities. The challenge, therefore, is not to make every development studies student an AI specialist, but to prepare graduates to interpret, question, govern and reshape AI-mediated development practice.
In my recent open access article, “Artificial intelligence, automation, and the future of development studies: rethinking teaching, learning, and graduate employability”, I argue that AI should be understood in three interconnected ways: as an object of critical development inquiry, as a force transforming teaching and assessment, and as a condition reshaping graduates’ professional futures.
AI is not a neutral tool
Development studies has a long tradition of questioning apparently neutral solutions to social problems. It asks who defines development priorities, whose knowledge counts, who controls resources and how the benefits and costs of change are distributed. The same questions must now be asked of AI.
AI systems are built within unequal global structures. Their development depends on large datasets, computational infrastructure, specialist expertise and considerable financial resources. These capacities are concentrated in a small number of corporations and countries. As AI systems spread into public administration, education, health, agriculture and development finance, they may create new dependencies for institutions that lack the resources or regulatory capacity to scrutinise them.
The appropriate question is therefore not only whether the technology works. We must also ask who owns the infrastructure, whose languages and knowledge are represented in the data, which communities are rendered visible or invisible, who is accountable when automated decisions cause harm, and what happens when systems designed elsewhere are introduced into institutionally and culturally different settings. These are not specialist computing questions. They are central questions about governance, power, justice and human development.
From producing answers to exercising judgement
Generative AI also challenges established assumptions about university learning. Systems can now produce essays, summarise literature, draft policy reports and generate apparently convincing explanations within seconds.
The immediate institutional response has understandably focused on academic integrity. However, plagiarism detection and prohibition cannot provide a sustainable educational strategy. The deeper issue is that polished written output can no longer be treated automatically as evidence of understanding.
Development studies education should therefore place greater emphasis on the intellectual processes behind an answer. Can students explain their reasoning? Can they evaluate conflicting evidence? Can they recognise weak assumptions? Can they identify what an AI-generated analysis overlooks? Can they connect general claims to a particular political, historical or institutional context?
This requires assessment to move from an exclusive focus on the final product towards greater attention to reasoning, verification and accountability. Written work will remain important, but it can be combined with oral defence, annotated research processes, reflective accounts, policy simulations, live briefs and collaborative problem-solving. The objective is not to design assessments that students cannot complete with AI. It is to design learning that requires them to exercise judgement about when, why and how AI should be used.
Beyond the search for fixed “future skills”
AI is also changing graduate employability. Repetitive, codifiable and text-intensive tasks are increasingly being automated or augmented. Yet this does not mean that human expertise is becoming irrelevant. It means that its value is shifting.
Development-related work frequently involves incomplete evidence, institutional constraints, competing interests and politically contested objectives. In these settings, technical proficiency alone is insufficient. Graduates must be able to interpret context, communicate across differences, evaluate ethical consequences and make defensible decisions under uncertainty.
The article describes this broader quality as adaptive capability. It comprises five closely connected capacities:
- Critical AI literacy: understanding what AI can and cannot do, and recognising bias, hallucination, opacity and weak evidence.
- Contextual judgement: assessing whether an AI-generated recommendation is appropriate for a particular community, institution or political setting.
- Ethical reflexivity: considering fairness, accountability, inclusion, representation and potential harm.
- Communicative and collaborative competence: explaining complex issues, negotiating across stakeholder groups and working effectively in interdisciplinary environments.
- Learning adaptability: revising knowledge and professional practice as technologies, institutions and labour markets continue to change.
These capabilities cannot be developed through a single workshop on prompt writing. They must be cultivated across the curriculum and through repeated opportunities for critical, applied and reflective learning.
A programme-wide response
An AI-responsive development studies curriculum does not require every module to become a technology module. It requires existing areas of study to examine how AI is changing the issues they already address.
Governance teaching can explore algorithmic decision-making and public accountability. Courses on poverty and inequality can investigate data bias, digital exclusion and unequal access to automated services. Labour modules can consider automation, platform work and worker agency. Research methods teaching can engage with the use of AI in literature mapping, coding and evidence analysis while emphasising verification and methodological transparency.
This programme-wide approach should be supported by staff development, clear assessment guidance, inclusive access and transparent institutional governance. It must also recognise substantial differences in infrastructure, affordability, linguistic representation and regulatory capacity across Global North and Global South institutions.
Renewing the purpose of development studies
The future of development studies does not depend on presenting itself as technologically fashionable. Its relevance lies in its ability to examine AI as part of wider struggles over knowledge, institutional power, inequality and human agency.
AI and automation do not displace the field’s traditional concerns. They intensify them.
Development studies should therefore prepare graduates who can do more than operate intelligent systems. They should be able to interpret them, question them, govern them and reshape their use in the service of more inclusive and equitable development.
The task is neither to embrace AI uncritically nor to defend established educational practices unchanged. It is to ensure that teaching, assessment and graduate preparation remain equal to the field’s critical mission in an age of intelligent systems.
Photo by Pixabay.
This blog is based on: Ekuma, K. (2026), “Artificial intelligence, automation, and the future of development studies: rethinking teaching, learning, and graduate employability”, Frontiers in Education, 11:1868968. Read the open-access article.
Author: Dr Kelechi Ekuma is a Senior Lecturer at the Global Development Institute, The University of Manchester. His work examines digital transformation, artificial intelligence, skills, human resource development and the future of education and work.
Note: This article gives the views of the author/academic featured and does not necessarily represent the views of the Global Development Institute as a whole.
Please feel free to use this post under the following Creative Commons license: Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0). Full information is available here.