OER, Generative AI, and Equity in African Distance Higher Education

Sixbert Sangwa; Placide Mutabazi

doi:10.20944/preprints202607.0109.v1

Submitted:

30 June 2026

Posted:

02 July 2026

You are already at the latest version

Abstract

African distance higher education is increasingly shaped by open educational resources (OER) and generative artificial intelligence, yet these agendas are often governed separately. This separation is problematic because OER and AI now meet in resource discovery, translation, adaptation, tutoring, feedback, assessment, accessibility, datafication, and knowledge production. This article develops an OER-AI Equity Framework for African distance higher education. It uses an integrative critical literature review, policy and document synthesis, and conceptual framework development. The synthesis draws on peer-reviewed scholarship, Open Praxis research, UNESCO guidance, African Union digital and AI strategies, connectivity data, accessibility standards, and literature on open pedagogy, digital equity, academic integrity, learner support, epistemic justice, and teacher agency. The analysis shows that equitable integration requires more than low-cost content and tool access. It requires seven interdependent commitments: infrastructural realism; epistemic justice and localisation; open pedagogy and learner agency; accessibility and learner support; assessment redesign and academic integrity; data protection and algorithmic accountability; and teacher agency with institutional policy coherence. The framework contributes a socio-technical, pedagogical, and governance-oriented lens for institutions, instructors, policymakers, quality assurance bodies, libraries, and researchers. It argues that African learners should not be positioned as passive consumers of imported content, platforms, or models, but as capable participants in open and AI-mediated knowledge systems.

Keywords:

open educational resources

;

generative artificial intelligence

;

distance education

;

digital equity

;

Africa

Subject:

Social Sciences - Education

Introduction

Open and distance higher education has long carried a democratic promise: to widen participation for learners whose work, geography, gendered responsibilities, disability, income, displacement, or institutional exclusion make conventional campus attendance difficult. In African higher education, that promise remains urgent as systems seek to expand enrolment, support lifelong learning, strengthen professional and teacher development, and serve learners who may combine study with employment, family care, community obligations, and uneven digital access. Yet the equity question in distance education can no longer be examined only through admission, devices, or programme availability. It must also be examined through the educational materials, platforms, algorithms, languages, assessment practices, and data relations through which learning is mediated.

Two developments now make this question particularly pressing. The first is the growth of open educational resources (OER), defined by UNESCO (2019) as learning, teaching, and research materials in any format that are in the public domain or openly licensed in ways that permit no-cost access, reuse, repurposing, adaptation, and redistribution. OER can reduce cost barriers, support local adaptation, widen access to learning materials, and enable open pedagogies in which students contribute to knowledge rather than only receive it. Reviews suggest that students using open textbooks generally perform as well as, and sometimes better than, those using commercial textbooks while saving money, although adoption depends on institutional culture, staff capacity, quality assurance, and support (Clinton & Khan, 2019; Hilton, 2016, 2019; Wiley & Hilton, 2018).

The second development is the rapid diffusion of generative artificial intelligence (AI) systems capable of producing text, images, summaries, translations, lesson plans, code, feedback, simulations, and conversational learning support. These tools have entered educational practice faster than many institutional policies, assessment systems, and teacher development structures can respond. UNESCO guidance calls for human-centred governance, regulation, capacity building, privacy protection, and attention to learners' rights (Miao et al., 2021; Miao & Holmes, 2023). The African Union's Continental Artificial Intelligence Strategy similarly frames AI as a development opportunity that must be governed through inclusion, human rights, dignity, capability building, contextual adaptation, and risk management (African Union, 2024).

The difficulty is that OER and generative AI are usually discussed in separate conversations. OER is often treated as a content, licensing, and affordability agenda; AI is often treated as a tutoring, productivity, automation, analytics, or assessment agenda. When separated in this way, each agenda becomes thinner than the educational realities it seeks to address. OER without AI may remain difficult to discover, translate, adapt, update, align, and make accessible in multilingual and resource-constrained systems. AI without openness may intensify dependence on proprietary platforms, opaque models, English-dominant corpora, data extraction, and commercial infrastructures that many African institutions do not control. When combined without equity safeguards, however, OER and AI may also create new harms: low-quality synthetic content, misattributed open materials, algorithmic bias, plagiarism disputes, surveillance, weakened teacher authority, and the further marginalisation of local knowledge traditions.

The African context makes these tensions sharp without making them uniform. The International Telecommunication Union (2024) estimated that 68% of the global population used the Internet in 2024, but only 38% of people in Africa were online, while low-income countries had an Internet use rate of 27%. These figures do not suggest a lack of African innovation or intellectual capacity. They show that any serious framework for OER and AI in African distance higher education must begin from infrastructural realism rather than abstract technological possibility. Learners may rely on mobile phones, shared devices, intermittent electricity, expensive data bundles, multilingual repertoires, asynchronous study rhythms, and local support networks. Institutions may have committed lecturers but constrained instructional design capacity, fragile learning platforms, limited digital library access, and uneven policy readiness. Equity therefore requires more than giving learners links to open resources or telling teachers to use AI responsibly.

This article develops a conceptual and policy-informed framework for integrating OER and generative AI in African distance higher education in ways that advance equity rather than merely accelerate digital delivery. The guiding argument is that OER-AI integration should be judged by its effects on participation, agency, dignity, knowledge production, accessibility, assessment validity, data rights, and teacher professionalism. The purpose is not to offer a universal technical recipe for all African institutions. Africa is not a single educational context, and distance higher education varies across national systems, languages, regulatory environments, institutional histories, pedagogical traditions, and resource conditions. The purpose is instead to provide a transferable equity framework that institutions and policymakers can adapt to their contexts.

The article is guided by four questions. First, how do OER and generative AI reshape access, affordability, participation, and knowledge production in African distance higher education? Second, what pedagogical, ethical, infrastructural, and governance risks emerge when OER and AI converge in resource-constrained distance education systems? Third, how can African institutions integrate OER and AI in ways that advance equity, epistemic justice, academic integrity, learner agency, and responsible innovation? Fourth, what conceptual framework can guide equitable OER-AI integration in African distance higher education? The article contributes the OER-AI Equity Framework for African Distance Higher Education, a seven-dimension model for institutional strategy, teaching practice, learner support, assessment redesign, data governance, and policy coherence.

Background and Related Literature

OER scholarship has moved from a narrow focus on cost reduction towards a broader discussion of openness, social justice, localisation, sustainability, and participation. Early adoption arguments often emphasised the financial burden of textbooks and the possibility that open materials could reduce or remove that burden. That argument remains important where students must choose among learning materials, transport, food, data, and family obligations. Meta-analyses and syntheses generally indicate that students using open textbooks perform as well as those using commercial textbooks while saving money (Clinton & Khan, 2019; Hilton, 2016, 2019). Student-facing studies also show that learners often perceive open textbooks as useful when quality and course alignment are clear, while equity-focused research links OER to reduced financial pressure for marginalised learners (Jhangiani & Jhangiani, 2017; Nusbaum et al., 2020). However, cost savings alone do not exhaust the meaning of openness. If open resources simply circulate dominant knowledge in cheaper form, they may widen access without challenging whose knowledge counts.

The social justice turn in open education is central to this article. Hodgkinson-Williams and Trotter (2018) argue that OER and open educational practices should be understood through redistributive, recognitive, and representational dimensions of justice. Redistribution concerns access to materials and resources; recognition concerns cultural, linguistic, and epistemic respect; representation concerns the ability of learners and educators to participate in decision-making and knowledge creation. Lambert (2018) similarly argues for a social justice aligned definition of open education that includes not only access but also recognitive and representational concerns. This matters because distance learners are often treated as recipients of content produced elsewhere. In African distance higher education, an equity-centred OER agenda must ask whether learners and teachers can adapt, critique, translate, author, and share resources that speak to local histories, professions, languages, and community problems.

Open pedagogy deepens this question. Cronin (2017) defines open educational practices as practices that include OER creation, reuse, open pedagogies, and open sharing of teaching practice. Wiley and Hilton (2018) describe OER-enabled pedagogy as teaching and learning practices made possible by the 5R permissions of OER: retain, reuse, revise, remix, and redistribute. A downloadable open textbook may reduce costs, but open pedagogy can reposition students as co-creators of renewable assignments, community resources, case studies, glossaries, practice guides, and public knowledge artefacts. For distance learners, this participatory dimension can reduce isolation and connect academic work to lived experience. Yet openness requires careful design. Public sharing can empower learners, but it can also expose them to reputational risk, extract unpaid labour, or privilege students with stronger digital skills, language capital, and connectivity.

Recent Open Praxis research confirms that OER remains an active concern in open and distance education and that regional adoption requires policy, infrastructure, and institutional support rather than isolated resource sharing (Khribi et al., 2026). At the same time, the journal's recent AI-focused work shows that generative AI is now reshaping teaching, assessment, agency, accessibility, and the verification of learning. Cefa et al. (2025) mapped early scholarly responses to ChatGPT in teaching, learning, and scholarship, while Bozkurt (2025) warned that generative and agentic AI can reconfigure student agency if adopted blindly. These debates are directly relevant to OER because generative AI now mediates how learners find, summarise, adapt, translate, cite, and produce learning resources.

The broader AI in education literature introduces opportunities and risks that cannot be separated from openness. Zawacki-Richter et al. (2019) found that much AI in higher education research was concentrated in computer science and STEM contexts and that educators were often underrepresented. Holmes and Tuomi (2022) warn that expectations about AI in education are frequently shaped by technical misunderstanding and narrow views of educational purpose. Kasneci et al. (2023) identify opportunities for large language models in personalised learning, accessibility, and feedback, but also note risks related to misinformation, overreliance, bias, privacy, and assessment integrity. These risks become more serious in distance education because learners may work asynchronously, use unsupported tools, leave extensive digital traces, and receive limited formative feedback.

Assessment is one of the clearest areas where OER and AI now intersect. Open pedagogy invites students to create reusable or public work; generative AI makes content generation faster and attribution more complex. Swiecki et al. (2022) argue that assessment in the age of AI requires rethinking what assessment is for, not only strengthening detection. Perkins et al. (2024) propose an AI Assessment Scale to clarify levels of permissible AI use, from no AI to full AI integration with human evaluation. For African distance higher education, the question is not simply whether students use AI. It is whether assessment designs value process evidence, local problem solving, oral explanation, critical reflection, ethical citation, and authentic application in contexts where learners may have unequal access to AI tools and connectivity.

Data governance is another area where OER and AI must be brought together. OER ecosystems already raise questions of copyright, attribution, licensing, platform ownership, and sustainability. AI systems add training data provenance, learner data collection, profiling, automated recommendations, hallucinated citations, and possible reproduction of social and linguistic bias. Slade and Prinsloo (2013) caution that learning analytics carries ethical dilemmas related to surveillance, consent, vulnerability, and power. Mittelstadt et al. (2016) map algorithmic ethics concerns around opacity, bias, responsibility, and the translation of values into technical systems. In distance education, where platforms mediate enrolment, content, communication, assessment, and support, data rights become part of educational equity rather than a peripheral compliance issue.

African distance higher education is not new to these tensions. Open universities, correspondence education, radio, print-based learning, mobile learning, blended support centres, and learning management systems have all been used to extend education beyond the conventional campus. Tait (2000) emphasised that learner support is not an optional supplement to open and distance learning but a core part of quality, retention, and success. Contemporary African digital education research also shows that students often access learning through constrained media ecologies. Loglo and Zawacki-Richter's (2023) systematic review of students' digital media use in African higher education found that learning management systems were mainly used for course delivery and were often accessed through weak internet-enabled mobile devices. Modise and Zawacki-Richter (2023) similarly positioned academic professional development as integral to online learning implementation in African open and distance teaching institutions.

This literature points to a precise gap. OER social justice frameworks explain redistribution, recognition, and representation. Digital equity frameworks explain systemic interactions among access, participation, and educational communication, including in open, distance, and digital education (Zawacki-Richter, 2026). AI ethics frameworks explain transparency, privacy, fairness, human oversight, and accountability (Miao & Holmes, 2023; UNESCO, 2021). Open pedagogy explains learner participation and renewable assignments. Yet none of these strands, by itself, is sufficient when OER and generative AI converge in African distance higher education. The distinct problem is that openness and AI operate together across content, pedagogy, assessment, accessibility, language, data, and institutional governance. The contribution of this article is therefore integrative: it develops a framework that treats OER-AI integration as a socio-technical, pedagogical, epistemic, and governance problem rather than as a licensing issue or a tool adoption issue alone.

Methods

Research method. This study used an integrative critical literature review, policy and document synthesis, and conceptual framework development design. The method was appropriate because the purpose was not to estimate an effect size, test an intervention, or conduct a systematic review of all studies on OER and AI. The purpose was to synthesise diverse but related bodies of scholarship and policy in order to construct a theoretically grounded framework for equity-centred OER-AI integration in African distance higher education. Integrative reviews are suitable when a field contains theoretical, empirical, and policy sources and when the goal is to generate conceptual understanding across streams of literature (Snyder, 2019; Torraco, 2005). Document analysis supported the examination of policy and institutional guidance, while framework development followed an iterative logic of concept identification, comparison, synthesis, and refinement (Bowen, 2009; Jabareen, 2009).

Research design or model. The design involved four connected phases. The first phase identified the conceptual problem: OER and generative AI are usually governed separately even though they increasingly converge in distance education practice. The second phase mapped literature and policy sources across OER, open pedagogy, AI in education, digital equity, African higher education, accessibility, academic integrity, learner support, data governance, epistemic justice, and teacher agency. The third phase synthesised recurring concepts, tensions, and implementation conditions. The fourth phase constructed and refined the OER-AI Equity Framework by asking which dimensions were necessary to connect openness, AI, equity, pedagogy, and governance without reducing the framework to a technology adoption checklist.

Data collection tools. The study used a structured document extraction matrix rather than surveys or interviews. The matrix recorded author or institutional source, year, source type, geographical scope, relevance to OER, relevance to AI, relevance to open and distance education, equity construct, governance issue, methodological contribution, and implication for framework development. For policy and standards sources, extraction focused on definitions, principles, obligations, implementation priorities, and risks. For peer-reviewed literature, extraction focused on theoretical constructs, findings, limitations, and relevance to African or Global South distance higher education. The matrix helped prevent the synthesis from becoming a narrative list of sources and supported comparison across evidence streams.

Sampling or research group. The unit of analysis was published literature, policy documents, standards, and institutional guidance rather than human participants. Sources were identified through purposive and iterative searching in Google Scholar, ERIC, publisher platforms, Open Praxis, the International Review of Research in Open and Distributed Learning, the Journal of Learning for Development, Computers and Education: Artificial Intelligence, Educational Technology Research and Development, UNESCO repositories, African Union repositories, ITU statistical reports, W3C standards, CAST guidance, and recognised open education sources. Search terms combined open educational resources, open pedagogy, open educational practices, generative AI, artificial intelligence in education, AI assessment, academic integrity, distance education, open and distance learning, digital equity, African higher education, Global South, epistemic justice, accessibility, data protection, learner support, teacher agency, and policy governance. Recent sources from 2020 to 2026 were prioritised for AI and policy issues, while older foundational works were retained where conceptually necessary. The final cited corpus contained 43 sources. Because the study was integrative and conceptual rather than systematic, no claim is made that the corpus represents an exhaustive database yield.

Inclusion criteria were: relevance to OER, open pedagogy, generative AI, AI in education, open and distance education, equity, Africa or the Global South, assessment, accessibility, data governance, learner support, teacher agency, or framework development; credibility as a peer-reviewed, official policy, standards, or authoritative technical source; and conceptual or practical relevance to higher education. Exclusion criteria were: promotional technology content, unverifiable sources, low-quality opinion pieces, unsubstantiated statistics, documents centred on proprietary product marketing, and sources not relevant to higher education or distance learning.

Research procedures. The synthesis proceeded through three analytical steps. First, sources were coded deductively using preliminary categories derived from the research questions: access, affordability, knowledge production, pedagogy, assessment, integrity, accessibility, data governance, teacher agency, and policy coherence. Second, inductive coding identified cross-cutting tensions, including open access without participation, AI productivity without pedagogical judgement, inclusion without infrastructure, localisation without quality support, accessibility without learner support, and governance principles without implementation capacity. Third, related codes were consolidated into seven framework dimensions. A dimension was retained only if it met three criteria: it appeared across more than one source cluster, it addressed a distinct equity risk or enabling condition, and it had practical implications for African distance higher education.

Validity and reliability measures. Trustworthiness was strengthened through source triangulation, policy-literature triangulation, explicit inclusion and exclusion criteria, preference for peer-reviewed and official sources, transparent documentation of framework construction, and negative case reasoning. Negative case reasoning asked how OER or AI could harm equity even when introduced with inclusive intentions. This was important because the framework sought neither technological enthusiasm nor technological rejection. The main limitation is that the framework has not yet been empirically validated through Delphi review, institutional case studies, learner experience research, or implementation trials. Its value at this stage is conceptual, analytical, and policy-oriented; future research should test its usability and explanatory value across different African distance education settings.

Table 1. Source selection and synthesis procedure.

Stage	Procedure	Output for the Synthesis
problem delimitation	Defined the convergence problem between OER, generative AI, African distance higher education, and equity.	four research questions and analytical boundaries
source identification	Used purposive and iterative searching across scholarly databases, publisher sites, Open Praxis, UNESCO, African Union, ITU, accessibility standards, and open education sources.	working corpus of peer-reviewed, policy, technical, and standards sources
eligibility screening	Retained sources that were credible, relevant to higher education, and useful for OER, AI, distance education, equity, governance, or framework development.	43 cited sources retained; promotional and unverifiable sources excluded
data extraction	Recorded source type, scope, OER relevance, AI relevance, equity construct, governance issue, and implications for framework development.	structured extraction matrix
synthesis and framework development	Applied deductive and inductive coding, compared source clusters, tested negative cases, and consolidated recurring equity conditions.	seven-dimension OER-AI Equity Framework

Note. The study is an integrative conceptual synthesis, not a systematic review. The cited corpus is reported for transparency rather than as an exhaustive search yield.

Findings and Discussion

The synthesis produced the OER-AI Equity Framework for African Distance Higher Education. The framework is not a maturity model, procurement guide, or accreditation checklist. It is an analytical and practical framework for asking whether the combination of OER and generative AI expands meaningful participation or merely introduces new forms of dependency. Its central claim is that equity in distance higher education is relational. It emerges through the interaction of learners, teachers, content, platforms, languages, assessment systems, institutional policies, support structures, and data infrastructures. A resource can be open and still exclusionary. A tool can be intelligent and still pedagogically shallow. A platform can be efficient and still extractive. The framework therefore requires institutions to treat OER-AI integration as a socio-technical and pedagogical project rather than a content automation strategy.

The framework adds to existing scholarship in three ways. First, it extends OER social justice work by showing how redistribution, recognition, and representation are altered when AI mediates access, adaptation, translation, feedback, and authorship. Second, it extends digital equity scholarship by locating equity not only in connectivity and participation, but also in open licensing, model governance, assessment validity, data protection, and epistemic agency. Third, it extends AI ethics and assessment literature by embedding AI governance within open education practice and African distance higher education realities. The framework therefore does not replace existing OER, AI ethics, or digital equity frameworks. It integrates their strongest insights around a convergence problem that has become unavoidable in open, distance, and digital education.

Table 2. OER-AI equity framework dimensions, equity questions, risks, and institutional responses.

Dimension	Equity Question	Key Risk if Weak	Institutional Response
infrastructural realism	Can learners access and use OER-AI resources under real study conditions?	Policies assume broadband, laptops, stable electricity, English fluency, and constant connectivity.	Design for mobile access, offline options, low bandwidth, asynchronous use, shared devices, and transparent cost assumptions.
epistemic justice and localisation	Whose knowledge is represented, adapted, cited, translated, and produced?	Open and AI-generated materials reproduce dominant languages, examples, histories, and professional assumptions.	Support local authorship, translation, contextual cases, community knowledge, and critical review of AI outputs.
open pedagogy and learner agency	Do learners participate as creators and reviewers, not only consumers?	Students complete extractive public tasks without consent, support, or meaningful ownership.	Use renewable assignments, co-created glossaries, local problem briefs, student consent, and safe sharing protocols.
accessibility and learner support	Are disabled, multilingual, rural, working, and first-generation learners fully included?	OER and AI resources are inaccessible, text-heavy, unaffordable, or unsupported.	Apply UDL, captions, transcripts, alternative text, assistive technology compatibility, helpdesk support, and learner orientation.
assessment and integrity	Does assessment value judgement, process, authenticity, and responsible AI use?	AI use is hidden, banned unrealistically, or permitted without criteria.	Redesign tasks, use AI-use declarations, process evidence, oral defence, authentic contexts, and transparent rubrics.
data and algorithmic accountability	How are learner data, prompts, outputs, platform logs, and analytics governed?	Learners and staff become data sources for opaque platforms without consent or accountability.	Adopt data minimisation, privacy notices, vendor review, human oversight, audit trails, and appeal mechanisms.
teacher agency and policy coherence	Are teachers equipped and trusted to exercise pedagogical judgement?	AI adoption deprofessionalises staff or leaves them unsupported.	Provide professional learning, workload recognition, communities of practice, local policy, and cross-unit governance.

Note. The framework is intended for adaptation across diverse African distance higher education contexts.

Infrastructural realism. The first dimension is infrastructural realism. Equity discussions often begin with noble aims, but distance learners experience educational technology through material constraints: device ownership, electricity, data costs, network reliability, platform usability, digital literacy, disability access, and time. ITU (2024) shows that global Internet use remains deeply unequal, with Africa below every other region in the proportion of people online. The implication is not that African distance education should avoid digital innovation. Rather, innovation must be designed for the conditions in which learners actually study. OER that requires continuous streaming, large downloads, or advanced devices may be legally open but practically closed. AI tools that require paid subscriptions, stable broadband, or advanced prompting skills may widen gaps among learners in the same programme.

Infrastructural realism requires institutions to make access conditions visible at the design stage. OER should be available in lightweight, printable, mobile-friendly, and offline formats where possible. AI-supported activities should include non-AI alternatives where access is unequal. Programme teams should estimate the real cost of participation, including data, device, printing, transport to support centres, and time for asynchronous study. This dimension also requires caution about platform dependence. A distance education system that relies on tools whose pricing, data practices, availability, and terms of service can change without institutional control is fragile. African Union (2020) digital transformation aspirations are best served when innovation is paired with infrastructure planning and learner protection.

Epistemic justice and localisation. The second dimension is epistemic justice. OER and AI are not neutral channels through which knowledge simply moves. They carry assumptions about whose knowledge is authoritative, whose language is standard, whose examples are normal, and whose problems are worth solving. Fricker (2007) describes epistemic injustice as a wrong done to people in their capacity as knowers. In higher education, this can occur when African learners encounter resources that treat their societies as case material rather than knowledge-producing contexts, or when AI systems generate responses based on corpora in which African scholarship, languages, and professional realities are underrepresented.

OER can help address this problem only if openness includes localisation, authorship, and recognition. Hodgkinson-Williams and Arinto's (2017) work on OER in the Global South shows that adoption is shaped by social, institutional, cultural, and technological conditions, not only licensing. Cox and Trotter (2017) similarly show that lecturers' OER adoption depends on institutional culture and perceived pedagogical value. In African distance higher education, localisation should not mean superficial substitution of names or examples. It should include rigorous adaptation of theories, cases, languages, datasets, and professional scenarios, together with proper attribution and quality review. AI can assist translation, summarisation, and adaptation, but human educators must judge whether outputs are accurate, contextually appropriate, and ethically usable. Mbembe's (2016) account of decolonising the university is relevant here because the issue is not simply content diversity; it is the structure of knowledge authority.

Open pedagogy and learner agency. The third dimension concerns learner agency. OER becomes more educationally significant when students can revise, remix, create, and redistribute knowledge under appropriate conditions. Open pedagogy can transform distance learners from isolated recipients into contributors to public goods, including local case repositories, multilingual glossaries, annotated readings, community guides, problem briefs, and practice-based resources. Generative AI complicates this opportunity. It can support brainstorming, translation, draft feedback, accessibility, and scaffolding for learners who struggle with academic language. Yet it can also reduce learning to prompt completion and output polishing if students are not required to explain, justify, revise, and situate their work.

Learner agency is therefore not achieved by giving students AI tools. It is achieved when students learn to evaluate AI outputs, compare them with evidence, cite sources responsibly, recognise limitations, and make accountable contributions. This is especially important in distance education, where learners may have rich professional and community experience that should not be overwritten by generic AI-generated answers. Institutions should design open assignments with consent, privacy, and choice. Not all learners can safely publish work openly. Some may face political, professional, religious, gendered, or family risks. Open pedagogy should therefore include graduated openness: private submission, class sharing, institutional repository sharing, public sharing, or anonymous contribution. Equity requires openness to be an opportunity, not coercive exposure.

Accessibility, inclusion, language, and learner support. The fourth dimension links OER-AI integration with accessibility and learner support. Digital resources are often celebrated for reach, but reach without accessibility reproduces exclusion. CAST (2024) presents Universal Design for Learning as a framework for designing for learner variability, while the World Wide Web Consortium (2023) provides accessibility guidance through WCAG 2.2. For distance higher education, this includes captions, transcripts, alternative text, accessible PDFs, keyboard navigation, screen-reader compatibility, readable structure, language clarity, flexible formats, and low-bandwidth alternatives. AI may support text-to-speech, summarisation, translation, and adaptive explanations, but it may also introduce errors, bias, or inaccessible interfaces.

Recent Open Praxis work on online and distance learners with visual impairments in relation to AI-assisted narration illustrates why learner experience must guide accessibility debates (Gülen et al., 2026). A tool that appears accessible from a technical perspective may still fail if it is inaccurate, difficult to navigate, unaffordable, or disconnected from assessment requirements. Learner support must therefore accompany technology. In African contexts, support should include orientation to OER licensing, AI literacy, digital study skills, library access, academic writing, assistive technologies, and routes for help when systems fail. Language deserves special attention. Generative AI may improve access through translation, but it may also flatten meaning, privilege high-resource languages, or generate unreliable translations in African languages with limited training data.

Assessment redesign and academic integrity. The fifth dimension concerns assessment. Generative AI has made it easier to produce plausible essays, summaries, lesson plans, code, and reflections. A narrow institutional response is to ban AI or intensify surveillance. Both responses are insufficient. Bans may be unrealistic where AI is embedded in common tools; surveillance can harm trust, privacy, and vulnerable students. The more defensible response is assessment redesign. Swiecki et al. (2022) argue that AI challenges inherited assumptions about assessment, while Perkins et al. (2024) provide a framework for clarifying levels of permissible AI use.

In an OER-AI environment, academic integrity should be understood as a design and learning issue. Assessment tasks should require local evidence, applied judgement, drafts, process logs, oral explanation, peer review, annotated source use, and reflection on AI assistance where permitted. Rubrics should distinguish between using AI as a tool and outsourcing judgement to AI. Students should be taught how to cite OER, acknowledge AI assistance, check sources, identify hallucinated references, and protect sensitive data. The goal is not to catch students after misconduct has occurred; it is to create assessment ecologies in which meaningful learning is more attractive, visible, and defensible than uncritical output generation. This is particularly important where unequal access to premium AI tools could make assessment unfair.

Data protection and algorithmic accountability. The sixth dimension is data and algorithmic accountability. Distance education systems already produce data through enrolment platforms, learning management systems, discussion forums, e-books, assessment submissions, helpdesk logs, and analytics. Generative AI adds prompts, uploaded documents, generated outputs, model interactions, and possible vendor retention of data. UNESCO's Recommendation on the Ethics of Artificial Intelligence and its education guidance place human rights, transparency, privacy, and accountability at the centre of AI governance (Miao & Holmes, 2023; UNESCO, 2021). The African Union (2024) similarly calls for an Africa-centric and inclusive AI approach.

For distance education, this means institutions should not treat AI tools as harmless productivity aids. They should review terms of service, prohibit the uploading of sensitive learner data into unapproved tools, clarify vendor relationships, document decisions, and maintain human oversight over consequential educational decisions. Learners should know when AI is used, what data are involved, and how they can challenge errors. Open licensing also requires governance. AI systems may generate resources that resemble copyrighted material, misattribute open works, or produce citations that do not exist. Institutions should teach staff and students to verify licences, retain source records, and distinguish public-domain, openly licensed, subscription-based, and AI-generated materials.

Teacher agency and institutional policy coherence. The seventh dimension concerns teacher agency and institutional coherence. AI discourse often imagines automation as a substitute for human teaching, while OER discourse sometimes assumes that materials circulate effectively once they are openly licensed. Both assumptions understate the role of teachers. Teachers select, adapt, explain, contextualise, question, support, assess, and care. They also mediate between global resources and local learners. UNESCO's AI competency frameworks emphasise the need to build teacher and student capacities rather than merely deploy tools (UNESCO, 2024a, 2024b).

Professional capacity is especially important where distance education institutions rely on part-time tutors, distributed centres, or lecturers already carrying heavy workloads. OER adaptation and AI integration require time: locating sources, checking licences, revising materials, testing accessibility, redesigning assessment, verifying AI outputs, and supporting learners. Without workload recognition, staff development, and communities of practice, institutions may shift the burden of innovation onto individual teachers. Institutional policy coherence means that OER, AI, assessment, data protection, accessibility, library services, learner support, and staff development should not be governed by isolated policies that contradict one another. Equity-centred integration requires cross-unit governance involving academic departments, libraries, ICT units, disability support, quality assurance, legal advisers, students, and senior leadership.

Implications

The framework has implications for institutions, instructors, libraries, policymakers, quality assurance bodies, and researchers. For institutions, OER-AI integration should be treated as a governed educational strategy rather than tool adoption. Policies should connect open licensing, AI use, accessibility, assessment, data protection, procurement, learner support, and staff development. For instructors, the framework points towards authentic, process-rich, locally situated, and transparent assignments. AI can support feedback, translation, and adaptation, but instructors remain responsible for disciplinary judgement, source verification, learning design, and pastoral-academic support in distance contexts.

For libraries and e-learning units, the framework expands the role of academic support. These units should help staff and students locate open resources, understand licences, verify AI-generated references, create accessible formats, and develop responsible AI literacy. For policymakers and regulators, the framework suggests that open education and AI governance should not be handled in separate silos. Guidance on data protection, AI in assessment, accessibility, teacher capacity, public-interest infrastructure, and open African knowledge repositories should be aligned. For researchers, the framework provides an agenda for empirical validation across varied African settings, including open universities, conventional universities expanding distance provision, private distance providers, regional partnerships, refugee education programmes, teacher education systems, and professional lifelong learning contexts.

Table 3. Implementation implications for major stakeholder groups.

Stakeholder Group	Primary Implication	Practical Action
institutions	Treat OER-AI integration as a governed educational strategy rather than tool adoption.	Create cross-unit policies connecting OER, AI, accessibility, assessment, data protection, procurement, learner support, and staff development.
instructors	Use AI and OER to deepen learning, not automate content delivery.	Design authentic tasks, verify AI outputs, adapt OER locally, require source transparency, and scaffold responsible student use.
learners	Develop agency as critical users and creators of open and AI-mediated knowledge.	Learn OER attribution, AI limitations, prompt ethics, data caution, process documentation, and reflective judgement.
libraries and e-learning units	Support discovery, licensing, quality review, accessibility, and responsible AI literacy.	Provide OER repositories, licence guidance, AI citation support, accessible formats, and staff and student training.
policymakers and regulators	Balance innovation with equity, rights, and contextual feasibility.	Issue guidance on open licensing, AI in assessment, data protection, accessibility, teacher capacity, and public-interest infrastructure.
researchers	Move from conceptual enthusiasm to empirical validation.	Conduct Delphi studies, institutional cases, learner experience research, and comparative African implementation studies.

Note. Actions are indicative and should be adapted to national regulation, institutional capacity, disciplinary context, and learner needs.

A final implication concerns quality. Equity should not be confused with lowering standards, and openness should not be confused with unrestricted circulation of unreviewed content. Poorly adapted OER and unverified AI outputs can harm learners precisely because they are inexpensive and easy to distribute. Equity-centred OER-AI integration therefore requires quality review, disciplinary expertise, citation discipline, accessibility testing, and feedback loops. The goal is not simply to make more resources available. It is to make educationally trustworthy resources and practices available in ways that learners can access, question, adapt, and build upon.

Limitations and Future Research

This article is a conceptual and policy-informed synthesis based on secondary sources. It does not report interviews, surveys, learning analytics, institutional case data, or learner outcome measures. The framework is therefore not presented as empirically validated. The African context is also diverse, and the framework cannot account for all differences among countries, institutions, languages, disciplines, disability contexts, learner populations, and regulatory systems. Further, generative AI is developing rapidly, so specific tools, risks, and institutional practices will change. The framework should therefore be treated as a living analytical guide rather than a fixed instrument.

Future research should validate the framework through expert Delphi studies, institutional case studies, comparative African policy analysis, learner experience research, and implementation studies in open and distance universities. Researchers should examine how learners with different connectivity, disability, language, gender, income, employment, and geographical conditions experience OER-AI integration. Studies should also test whether assessment redesign strategies reduce misconduct concerns while strengthening learning. Finally, African scholars and institutions should lead research on local-language AI, open African knowledge repositories, community-authored OER, public-interest AI infrastructures, and institutional models that protect teacher agency while expanding learner support.

Conclusion

This article has argued that OER and generative AI should be integrated in African distance higher education through an explicit equity framework. The central contribution is the OER-AI Equity Framework for African Distance Higher Education, which brings together infrastructural realism, epistemic justice, open pedagogy, accessibility, assessment redesign, data accountability, and teacher agency. The framework responds to a gap in the literature: OER and AI are often discussed separately even though they now meet directly in resource design, translation, tutoring, feedback, assessment, learner support, data governance, accessibility, and institutional policy. Treating them separately weakens both agendas. Treating them together without equity safeguards risks amplifying old inequalities in more efficient forms.

The framework's central message is that openness should be judged by practical participation, not by licence status alone, and that AI governance should be pedagogical as well as technical. A resource is not genuinely equitable if learners cannot access it, understand it, adapt it, trust it, or connect it to their contexts. An AI tool is not educationally responsible if it reduces students to data sources, automates judgement, hides its assumptions, or weakens teacher agency. For African distance higher education, the future of OER and AI should not be framed as a choice between enthusiasm and rejection. The better question is whether institutions can organise openness, AI, pedagogy, infrastructure, data, and policy around learner dignity, African knowledge production, accountable assessment, and human educational judgement.

Ethics and consent: This study used publicly available literature, policy documents, standards, and secondary sources only. It did not involve human participants, personal data collection, confidential institutional records, or intervention with learners or staff. Ethics approval was therefore not required.

Data accessibility statement: All sources analysed in this article are publicly available and cited in the reference list. No primary dataset was generated or analysed.

Funding information: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Competing interests: The author declares no competing interests.

Author contributions

The authors were responsible for conceptualisation, methodology, investigation, formal analysis, writing of the original draft, and writing, review, and editing.

Generative AI assistance statement: This manuscript was drafted, reviewed, edited, and refined with assistance from OpenAI's ChatGPT (GPT-5.5 Thinking, June 2026) for language refinement, structural improvement, and editorial review. The human author critically reviewed, revised, and validated all AI-assisted content to ensure academic rigour, source accuracy, ethical compliance, and contextual appropriateness. The author also assessed and addressed potential biases in AI-assisted wording. The final content, analysis, conclusions, and scholarly integrity of the manuscript are the sole responsibility of the human author.

Acknowledgments

No acknowledgements.

References

African Union. The digital transformation strategy for Africa (2020-2030). 2020. Available online: https://au.int/sites/default/files/documents/38507-doc-DTS_for_Africa_2020-2030_English.pdf.
African Union. Continental artificial intelligence strategy: Harnessing AI for Africa's development and prosperity. 2024. Available online: https://au.int/sites/default/files/documents/44004-doc-EN-_Continental_AI_Strategy_July_2024.pdf.
Bowen, G. A. Document analysis as a qualitative research method. Qual. Res. J. 2009, 9(2), 27–40. [Google Scholar] [CrossRef]
Bozkurt, A. Algorithmically manufactured minds: Generative and agentic AI in a time of post-truth, reconfiguration of student agency and death of critical pedagogy. Open Prax. 2025, 17(2), 206–210. [Google Scholar] [CrossRef]
CAST. Universal design for learning guidelines 3.0. 2024. Available online: https://udlguidelines.cast.org/.
Cefa, B.; Macgilchrist, F.; ElGamal, H.; Bai, J. Y. H.; Zawacki-Richter, O.; Loglo, F. S. Y. Responses to the initial hype: ChatGPT supporting teaching, learning, and scholarship? Open Prax. 2025, 17(2), 227–250. [Google Scholar] [CrossRef]
Clinton, V.; Khan, S. Efficacy of open textbook adoption on learning performance and course withdrawal rates: A meta-analysis. AERA Open 2019, 5(3), 1–20. [Google Scholar] [CrossRef]
Cox, G.; Trotter, H. Factors shaping lecturers' adoption of OER at three South African universities. In Adoption and impact of OER in the Global South. African Minds, International Development Research Centre, & Research on Open Educational Resources for Development.; Hodgkinson-Williams, C., Arinto, P. B., Eds.; 2017. [Google Scholar] [CrossRef]
Cronin, C. Openness and praxis: Exploring the use of open educational practices in higher education. Int. Rev. Res. Open Distrib. Learn. 2017, 18(5), 15–34. [Google Scholar] [CrossRef]
Fricker, M. Epistemic injustice: Power and the ethics of knowing; Oxford University Press, 2007. [Google Scholar] [CrossRef]
Gülen, S. K.; Çöpgeven, N. S.; Tuna Büyükköse, G.; Erdoğdu, E.; Uçar, H.; Aydemir, M.; Doğuş, M.; Arı, S. Exploring the experiences of online and distance learners with visual impairments in regard to AI-assisted narration. Open Prax. 2026, 18(2), 365–380. [Google Scholar] [CrossRef]
Hilton, J. Open educational resources and college textbook choices: A review of research on efficacy and perceptions. Educ. Technol. Res. Dev. 2016, 64, 573–590. [Google Scholar] [CrossRef]
Hilton, J. Open educational resources, student efficacy, and user perceptions: A synthesis of research published between 2015 and 2018. Educ. Technol. Res. Dev. 2019, 68, 853–876. [Google Scholar] [CrossRef]
Hodgkinson-Williams, C.; Arinto, P. B. (Eds.) Adoption and impact of OER in the Global South. In African Minds, International Development Research Centre, & Research on Open Educational Resources for Development; 2017. [Google Scholar] [CrossRef]
Hodgkinson-Williams, C. A.; Trotter, H. A social justice framework for understanding open educational resources and practices in the Global South. J. Learn. Dev. 2018, 5(3), 204–224. Available online: https://jl4d.org/index.php/ejl4d/article/view/312. [CrossRef]
Holmes, W.; Tuomi, I. State of the art and practice in AI in education. Eur. J. Educ. 2022, 57(4), 542–570. [Google Scholar] [CrossRef]
International Telecommunication Union. Measuring digital development: Facts and figures 2024. 2024. Available online: https://www.itu.int/itu-d/reports/statistics/facts-figures-2024/.
Jabareen, Y. Building a conceptual framework: Philosophy, definitions, and procedure. Int. J. Qual. Methods 2009, 8(4), 49–62. [Google Scholar] [CrossRef]
Jhangiani, R. S.; Jhangiani, S. Investigating the perceptions, use, and impact of open textbooks: A survey of post-secondary students in British Columbia. Int. Rev. Res. Open Distrib. Learn. 2017, 18(4), 172–192. [Google Scholar] [CrossRef]
Kasneci, E.; Sessler, K.; Küchemann, S.; Bannert, M.; Dementieva, D.; Fischer, F.; Gasser, U.; Groh, G.; Günnemann, S.; Hüllermeier, E.; Krusche, S.; Kutyniok, G.; Michaeli, T.; Nerdel, C.; Pfeffer, J.; Poquet, O.; Sailer, M.; Schmidt, A.; Seidel, T.; Kasneci, G. ChatGPT for good? On opportunities and challenges of large language models for education. Learn. Individ. Differ. 2023, 103, Article 102274. [Google Scholar] [CrossRef]
Khribi, M. K.; Belhadj, H.; Tlili, A.; Sellami, A.; Jemni, M. Open educational resources in the Gulf Cooperation Council (GCC) countries: An integrated systematic review. Open Prax. 2026, 18(2), 324–346. [Google Scholar] [CrossRef]
Lambert, S. R. Changing our (dis)course: A distinctive social justice aligned definition of open education. J. Learn. Dev. 2018, 5(3), 225–244. [Google Scholar] [CrossRef]
Loglo, F. S.; Zawacki-Richter, O. Learning with digital media: A systematic review of students' use in African higher education. J. Learn. Dev. 2023, 10(1), 1–23. [Google Scholar] [CrossRef]
Mbembe, A. J. Decolonizing the university: New directions. Arts Humanit. High. Educ. 2016, 15(1), 29–45. [Google Scholar] [CrossRef]
Miao, F.; Holmes, W.; Huang, R.; Zhang, H. AI and education: Guidance for policy-makers; UNESCO, 2021; Available online: https://unesdoc.unesco.org/ark:/48223/pf0000376709.
Miao, F.; Holmes, W. Guidance for generative AI in education and research; UNESCO, 2023; Available online: https://unesdoc.unesco.org/ark:/48223/pf0000386693.
Mittelstadt, B. D.; Allo, P.; Taddeo, M.; Wachter, S.; Floridi, L. The ethics of algorithms: Mapping the debate. Big Data Soc. 2016, 3(2), 1–21. [Google Scholar] [CrossRef]
Modise, M. P.; Zawacki-Richter, O. Professional development of academics for the implementation of online learning in African open and distance teaching institutions. Int. J. E-Learn. Distance Educ. 2023, 37(2). [Google Scholar] [CrossRef]
Nusbaum, A. T.; Cuttler, C.; Swindell, S. Open educational resources as a tool for educational equity: Evidence from an introductory psychology class. Front. Educ. 2020, 4, 152. [Google Scholar] [CrossRef]
Perkins, M.; Furze, L.; Roe, J.; MacVaugh, J. The Artificial Intelligence Assessment Scale (AIAS): A framework for ethical integration of generative AI in educational assessment. J. Univ. Teach. Learn. Pract. 2024, 21(6). [Google Scholar] [CrossRef]
Slade, S.; Prinsloo, P. Learning analytics: Ethical issues and dilemmas. Am. Behav. Sci. 2013, 57(10), 1510–1529. [Google Scholar] [CrossRef]
Snyder, H. Literature review as a research methodology: An overview and guidelines. J. Bus. Res. 2019, 104, 333–339. [Google Scholar] [CrossRef]
Swiecki, Z.; Khosravi, H.; Chen, G.; Martinez-Maldonado, R.; Lodge, J. M.; Milligan, S.; Selwyn, N.; Gašević, D. Assessment in the age of artificial intelligence. Comput. Educ. Artif. Intell. 2022, 3, 100075. [Google Scholar] [CrossRef]
Tait, A. Planning student support for open and distance learning. Open Learn. J. Open Distance E-Learn. 2000, 15(3), 287–299. [Google Scholar] [CrossRef]
Torraco, R. J. Writing integrative literature reviews: Guidelines and examples. Hum. Resour. Dev. Rev. 2005, 4(3), 356–367. [Google Scholar] [CrossRef]
UNESCO. Recommendation on open educational resources (OER). 2019. Available online: https://unesdoc.unesco.org/ark:/48223/pf0000383205.
UNESCO. Recommendation on the ethics of artificial intelligence. 2021. Available online: https://unesdoc.unesco.org/ark:/48223/pf0000381137.
UNESCO. AI competency framework for teachers. 2024a. Available online: https://unesdoc.unesco.org/ark:/48223/pf0000391104.
UNESCO. AI competency framework for students. 2024b. Available online: https://unesdoc.unesco.org/ark:/48223/pf0000391105.
Wiley, D.; Hilton, J. L., III. Defining OER-enabled pedagogy. Int. Rev. Res. Open Distrib. Learn. 2018, 19(4), 133–147. [Google Scholar] [CrossRef]
World Wide Web Consortium. Web content accessibility guidelines (WCAG) 2.2. 2023. Available online: https://www.w3.org/TR/WCAG22/.
Zawacki-Richter, O. Towards a systems-based framework for digital educational equity in open, distance, and digital education. Open Prax. 2026, 18(2), 381–397. [Google Scholar] [CrossRef]
Zawacki-Richter, O.; Marín, V. I.; Bond, M.; Gouverneur, F. Systematic review of research on artificial intelligence applications in higher education: Where are the educators? Int. J. Educ. Technol. High. Educ. 2019, 16, 39. [Google Scholar] [CrossRef]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

© 2026 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

Copyright: This open access article is published under a Creative Commons CC BY 4.0 license, which permit the free download, distribution, and reuse, provided that the author and preprint are cited in any reuse.