ORIGINAL RESEARCH ARTICLE
Generative AI as a partner for teachers in building personalised learning paths for students with ease in Tanzania
Juliana Kamaghe*
Mathematics & Information and Communication Technology Department, The Open University of Tanzania, Tanzania
Received: 21 June 2025; Revised: 19 October 2025; Accepted: 28 October 2025; Published: 10 February 2026
This study examines how generative artificial intelligence (AI) can assist secondary school teachers in Tanzania to create personalised learning paths more efficiently and effectively. Many educators face overcrowded classrooms and limited resources, making it challenging to meet the diverse needs of their students. To address this, 120 Dar es Salaam and Dodoma teachers tested AI-driven tools like ChatGPT and Grok for lesson planning, assessments and adaptive content delivery. The results indicated significant improvements in student engagement and academic performance while reducing teacher workload. Teachers found these AI tools intuitive and beneficial, especially for customising instruction and saving time. However, challenges such as inadequate training and infrastructure continue to pose significant obstacles, particularly in rural areas. The study concludes that generative AI offers a scalable and inclusive solution for enhancing teaching and learning when paired with proper support. It recommends strategic investments in professional development and digital infrastructure to fully realise generative AI’s educational potential and address existing equity gaps across Tanzanian schools.
Keywords: Generative AI in education; personalised learning; digital teachers; AI integration in secondary education; educational technology
*Corresponding author. Email: julianakamaghe@gmail.com
Research in Learning Technology 2026. © 2026 J. Kamaghe. Research in Learning Technology is the journal of the Association for Learning Technology (ALT), a UK-based professional and scholarly society and membership organisation. ALT is registered charity number 1063519. http://www.alt.ac.uk/. This is an Open Access article distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), allowing third parties to copy and redistribute the material in any medium or format and to remix, transform, and build upon the material for any purpose, even commercially, provided the original work is properly cited and states its license.
Citation: Research in Learning Technology 2026, 34: 3594 - http://dx.doi.org/10.25304/rlt.v34.3594
Introduction
The rapid evolution of artificial intelligence (AI) has paved the way for innovative pedagogical approaches in education (Miao & Holmes, 2021). One of the most promising applications of AI in education is the development of personalised learning paths that adapt teaching materials and instructional methods to meet the unique needs of individual students. This article explores the potential of leveraging generative AI to create personalised learning experiences in Tanzanian classrooms, serving as a practical assistant to teachers while aiming to improve student performance and engagement (Arias-Flores et al., 2025; Holmes, 2020).
The integration of AI into education has ushered in transformative possibilities, particularly through the advent of generative AI tools that create content, adapt resources and support human efforts in unprecedented ways. As classrooms grow increasingly diverse and student needs become more complex, teachers face mounting challenges in delivering personalised learning experiences that cater to individual strengths, weaknesses and interests (Arias-Flores et al., 2025; Miao & Holmes, 2021). Generative AI – encompassing technologies like OpenAI’s GPT-5 models, DALL-E for image generation and xAI’s SuperGrok – offers a promising avenue to address these challenges by partnering with educators rather than replacing them. This article explores how such tools can simplify the creation of tailored learning paths, empowering teachers to enhance their efficacy and improve student outcomes in an era of technological innovation and evolving educational demands (Kasneci et al., 2023; Zhao et al., 2024).
Background and rationale
Personalised learning, a pedagogical approach that customises education to individual student profiles, has long been recognised as a driver of engagement and academic success (Bernackiet al., 2021; Lin et al., 2024). However, its implementation is hindered by substantial obstacles, including time constraints, limited resources, and the diverse needs of learners (Lin et al., 2024; Zhao et al., 2024). Teachers often struggle to design adaptive lesson plans or provide individualised feedback within traditional classroom settings (Wang et al., 2025). The rise of AI in education has introduced tools to alleviate these burdens, with generative AI emerging as a standout innovation. Technologies like GPT-4 and DALL-E (Arias-Flores et al., 2025; Ranganai et al., 2022) can generate text, visuals and interactive content, while platforms like Grok, developed by xAI, provide real-time assistance tailored to specific queries (Dwivedi et al., 2023). Studies highlight their potential in automating administrative tasks and enhancing content delivery (Holmes, 2023), yet their role in supporting teachers’ creative and pedagogical agency remains underexplored.
Global trends in generative AI in education
Integrating AI in education reshapes traditional pedagogical models through personalised learning pathways. Numerous studies have established that AI-enhanced learning environments significantly improve academic outcomes, learner engagement and knowledge retention (Holmes, 2020; Marques-Cobeta, 2024). AI algorithms, particularly those enabling adaptive learning, help tailor educational experiences to individual learners by analysing patterns in performance, identifying strengths and weaknesses, and customising content delivery accordingly (Alam, 2023). Generative AI tools, such as ChatGPT and other large language models, are increasingly used to create dynamic instructional materials, generate personalised quizzes and support inquiry-based learning, facilitating more engaging and inclusive educational experiences (Laak & Aru, 2024).
Moreover, generative AI addresses key limitations of traditional education systems by supporting students whose rigid, standardised methods may otherwise marginalise them. Research indicates that learners interacting with AI-generated resources tend to display increased perseverance in complex subjects and higher intrinsic motivation (Aleven et al., 2023). Real-time analytics and intelligent tutoring systems (ITS) provide educators with actionable feedback, allowing for continuous refinement of learning strategies and improved academic support tailored to each student (Huang et al., 2023).
Relevance to the Tanzanian educational context
In Tanzania, as in many sub-Saharan African countries, challenges such as overcrowded classrooms, scarce teaching materials and heterogeneity in student needs hinder effective teaching and learning (Bighamet al., 2017; Said, 2018). These structural limitations make the case for personalised, AI-driven instruction powerful. While infrastructure gaps remain a barrier, early evidence suggests that even low-resource contexts can benefit from targeted AI interventions that optimise teacher–student interaction and support differentiated instruction (Maritim & Mushi, 2012; Mnyawami et al., 2022; Sedoyeka, 2012).
The adoption of generative AI tools in Tanzanian schools holds promise for transforming education by supplementing teacher capacity and offering individualised learning pathways. Applying these models in Tanzania could help bridge educational inequalities, foster digital literacy and prepare learners to thrive in an increasingly knowledge-driven economy (Ishengoma & John, 2024; Mtebe & Raphael, 2018; Ponera & Madila, 2024). Despite the proliferation of AI research in education, much of the literature focuses on AI as a standalone solution, potentially supplanting human educators rather than complementing them. For instance, Yang et al., (2020). emphasises machine learning’s predictive power in modelling complex systems. Tellman et al. (2021) showcase sate applications I for large-scale applications, but neither addresses teacher-centred collaboration. Similarly, studies on teachers remain central to education, needing tools that amplify their efforts rather than replace them (Cukurova et al., 2020; Molenaar, 2021). The ease of use and the integration of generative AI into daily teaching practices also receive insufficient attention, limiting its potential in resource-constrained settings.
This study proposes generative AI as a collaborative partner for teachers, enabling them to build personalised learning paths more easily and efficiently. The significance of this approach lies in its dual benefits: for teachers, generative AI reduces workload and sparks creativity by automating routine tasks and suggesting innovative ideas; for students, it delivers individualised support that aligns with their learning pace and preferences (Xie et al., 2019; Zhai et al., 2021). As education systems worldwide grapple with equity and scalability challenges, this partnership offers a human-centred solution that bridges technological potential with classroom realities, promising a more inclusive and effective learning environment.
Literature review
Personalised learning in education
Personalised learning aims to adapt educational experiences to individual student needs, preferences and abilities, moving beyond the one-size-fits-all model of traditional education. Vygotsky’s zone of proximal development provides a theoretical basis, suggesting that learning is most effective when scaffolded to a student’s current level. Traditional methods, such as differentiated instruction, have sought to achieve this by varying content, processes or products based on student readiness (Tomlinson, 2016). However, these approaches face significant limitations. Research by Thomas (2023) highlights that teachers struggle to implement personalisation at scale due to time constraints and resource shortages, with only 20% of surveyed educators reporting consistent success. Similarly, Marques-Cobeta (2024) notes that manual differentiation often fails to address diverse learning paces, leaving gaps in student engagement and achievement. These challenges underscore the need for tools that simplify and enhance personalised teaching, particularly in resource-constrained settings.
Generative AI and teacher support
The role of generative AI in education has evolved from automating tasks to supporting human instructors, a shift increasingly documented in recent literature. Holmes (2020) provide a seminal review, noting that early AI systems, such as ITS, aimed to replace teachers but often lacked the nuance of human judgement. More recent studies emphasise collaboration over substitution. For example, Octavio et al. (2024) describe AI as a ‘teacher’s assistant’, capable of handling repetitive tasks (e.g. grading) to free educators for higher-order instruction. Molenaar (2022) extends this idea, showing that hybrid human–AI systems improve student outcomes by 12% in adaptive learning environments. Specific to generative AI, Baidoo-Anu and Ansah (2023) found that teachers using ChatGPT for lesson planning reported a 25% reduction in preparation time, enhancing their focus on student interaction. Similarly, Selwyn (2021) highlights generative AI’s role in scaffolding teacher creativity, though he warns of risks of over-reliance. These findings suggest a growing consensus: generative AI’s value lies in empowering, not replacing, educators, a perspective this study builds upon. To successfully implement generative AI in Tanzanian classrooms, it is essential to integrate these technologies within existing educational frameworks. Many schools already utilise Learning Management System (LMSs) for administrative tasks and content distribution, providing a ready-made platform for AI deployment.
Research gap
Despite these advances, significant gaps remain in understanding how generative AI can support teachers in personalised learning with ease and agency. Much existing research focuses on student-facing AI applications (Tellman et al., 2021) or technical performance (Mosavi et al., 2018), sidelining the teacher’s experience. Holmes et al. (2022) note that usability, a critical factor in technology adoption, is underexplored, with only 10% of AI education studies addressing teacher perceptions of ease. Moreover, teacher agency, defined as the capacity to shape instructional tools (Filgueiras et al., 2015), is often overlooked in AI design, risking systems that dictate rather than assist. Studies by Xia et al. (2023) call for intuitive interfaces to bridge this gap, yet few have tested generative AI in this context. Regional disparities persist; for instance, Nzeng’e et al. (2021) highlight that in East Africa, AI adoption lags due to infrastructure and training deficits, limiting its personalisation potential.
Research methodology
This study aimed to investigate how generative AI supports Tanzanian teachers in creating personalised learning paths for students with ease, addressing the unique challenges and opportunities within the country’s education system. Tanzania’s education landscape, shaped by rapid population growth, resource constraints and a push for digital transformation, provides a critical context for exploring the potential of AI. With over 60 million people and a significant rural–urban divide, the system faces issues such as overcrowded classrooms, limited teacher training and uneven access to technology. This methodology employs a robust, context-sensitive approach to assess the usability and impact of generative AI tools, offering insights that can inform educational policy and practice in Tanzania and similar settings.
Research design
This study adopts a mixed-methods approach, combining qualitative and quantitative techniques to comprehensively evaluate generative AI’s role in personalised teaching. A case study design anchors the research within specific Tanzanian educational settings, capturing real-world dynamics. Surveys complement this by gathering broader perceptions from teachers, while an experimental design tests the practical impact of AI tools on teaching and learning outcomes, aligning with the mixed-methods framework by Creswell and Clark (2017). This triangulation enhances validity, addressing teachers’ subjective experiences and objective student performance metrics.
Study participants
The study involves 120 educators from diverse Tanzanian educational contexts, including public secondary schools in Dar es Salaam and Dodoma. Participants represent a mix of disciplines (e.g. science, mathematics, languages) to ensure broad applicability, mirroring the multi-disciplinary focus of Tanzania’s Education Sector Development Plan (2021–2026). Teachers are selected based on their willingness to adopt AI tools and varying levels of technological familiarity, reflecting the digital divide noted by Sedoyeka (2012). This sample size balanced depth and feasibility, consistent with recommendations for educational case studies while capturing urban–rural disparities critical to Tanzania’s context.
Tools evaluated
Two generative AI platforms, namely ChatGPT (OpenAI) and Grok (xAI), were employed and chosen for their accessibility and versatility in educational tasks. ChatGPT generates lesson plans, quizzes and student feedback, leveraging its natural language capabilities, while Grok, with its focus on reasoning, assists in creating problem-solving activities tailored to student needs. These tools align with UNESCO’s (2022) call for AI solutions that enhance teaching efficiency. In Tanzania, where internet access is improving, these cloud-based platforms are viable, though offline alternatives may be explored for rural settings.
Data collection
Data collection focused on public schools where teachers completed pre- and post-intervention surveys adapted from the Technology Acceptance Model, assessing ease of use, time saved and satisfaction with AI tools. Student performance metrics, including engagement and grades, were gathered from class records before and after AI implementation, following methodologies outlined in Holmes et al. (2020). In the context of Tanzania, where classroom sizes average 40–60 students, these metrics highlight the scalability of AI.
Data analysis
Qualitative data from teacher feedback were analysed using thematic coding, guided by Braun and Clarke’s (2006) framework. Quantitative data from survey scores and student outcomes were analysed using paired t-tests to compare pre- and post-intervention differences. Effect sizes were calculated to assess practical significance, addressing the need for impactful interventions in resource-limited settings (Kiunsi, 2013). This dual analysis ensures that the findings are statistically robust and contextually relevant, bridging global AI research (Mosavi et al., 2018) with Tanzania’s educational system priorities.
Study design block diagram
The workflow of the study is depicted in a block diagram. Figure 1 illustrates the input block, where student profiles (lesson goals) are fed into the AI system based on the generated prompts. The decision block indicates which generative AI tool (such as ChatGPT or Grok) will process the inputs to produce outputs like lesson plans and teaching materials. This system generates personalised material tailored to students’ learning outcomes. Lastly, the evaluation block facilitates surveys, logs and analysis of student data to assess impact and refine AI usage based on feedback. This design, inspired by system dynamics in educational technology, visually represents the partnership between AI and teachers.
Source: Researcher.
Ethical considerations
Ethical protocols are paramount, given Tanzania’s emphasis on equity and data protection in education. Informed consent was obtained from all participants who were informed about the study’s purpose and voluntary nature. Data privacy was ensured through anonymisation and secure storage on encrypted servers, which was critical in a country with emerging cybersecurity frameworks. AI bias, such as cultural insensitivity in generated content, was mitigated by reviewing outputs for relevance to Tanzania’s curriculum and Kiswahili-language needs, as Kitalima (2024) emphasised.
Results and discussion
The study surveyed 120 Dar es Salaam and Dodoma secondary school teachers, offering a comprehensive demographic overview. Female participants comprised a slight majority at 58%, while males represented 42%, reflecting relatively balanced gender participation. Age-wise, the largest group (62%) was between 30 and 40 years, indicating a predominance of mid-career professionals. Teachers aged 41–50 comprised 22%, while those younger than 30 or older than 50 accounted for the remaining 16%, highlighting a modest presence of both early-career and experienced educators.
Regarding professional experience, nearly half (48%) of the teachers had been teaching for 5–10 years, suggesting a workforce with solid classroom experience. Additionally, 30% had more than a decade of experience, while 22% were relatively new, with less than 5 years in the profession. Regarding qualifications, 29% hold a diploma in teacher education, most of them (36%) have bachelor’s degrees, and 35% possess postgraduate credentials, including master’s degrees and educational diplomas, suggesting a well-qualified teaching cohort, as shown in Figure 2.
Geographically, most respondents (70%) were from urban schools, with 30% serving in rural institutions in the Dar es Salaam and Dodoma regions. This urban-dominated sample provides valuable insight into the teacher demographics across different settings in Tanzania’s secondary education sector.
Comparison of traditional versus AI-driven personalised learning
The research compared the traditional learning for secondary schools in Tanzania, and it was observed that the data presented below, gathered from 120 secondary school teachers across Tanzania, reveal a substantial perceived improvement in multiple educational dimensions when comparing traditional teaching methods to AI-driven personalised learning systems.
Teachers overwhelmingly rated AI-enhanced approaches higher across all measured categories. Remarkably, timely feedback and assessment received a mean score of 4.6, compared to 2.6 under traditional methods, indicating that real-time analytics and AI feedback mechanisms significantly enhance formative assessment processes. Similarly, adaptability to individual learning needs improved from 2.5 to 4.5, underscoring AI’s strength in catering to diverse learner profiles, as supported by Thomas (2023). Moreover, the data reflect a dramatic improvement in student engagement and motivation, both critical to learning retention. Teachers noted a shift in classroom dynamics, where students became more autonomous and interested when interacting with generative AI content such as adaptive quizzes or scenario-based simulations (Holmes, 2023), as shown in Figure 3.
Figure 3. Comparison of traditional versus AI-driven personalised learning.
Interestingly, teacher workload reduction scored 4.0 for AI-assisted models, compared to 2.2 in traditional settings, suggesting that automation of administrative tasks (e.g. grading and progress tracking) allows educators to focus more on instruction and mentoring (Woolf et al., 2013). This table highlights the potential of AI-driven approaches to address the unique challenges faced by the Tanzanian education system by ensuring that content and instruction are continuously adapted to meet learner needs.
Current status of teacher readiness
A critical barrier to the successful integration of AI in educational settings is the preparedness and confidence of educators to adopt these new technologies. Seventy-six percent of teachers pointed out that teacher readiness and attitudes significantly influence their intention to integrate AI into their curriculum, while 26% indicated that they are not influenced by it. However, the lack of adequate training and familiarity with AI often results in hesitance to adopt these technologies in everyday teaching practices.
For instance, research shows that 56% of teachers demonstrated readiness for AI education, and while 46% were optimistic regarding the potential of AI, insufficient training in technological and pedagogical content knowledge (TPACK) remains a significant challenge, especially in rural areas of Tanzania. Similar findings from investigations in diverse educational contexts suggest that extensively improved professional development programmes must cover the practical aspects of using AI tools in classrooms.
Teachers’ perceptions of generative AI
The results show significant differences between teachers using generative AI tools and those who do not. Teachers with AI experience rated ease of use significantly higher (M = 4.3, standard deviation [SD] = 0.6) than those without experience (M = 3.1, SD = 0.9), t (118) = 4.62, p < 0.001. This suggests that teachers find generative AI intuitively operable once engaged, likely influenced by user-friendly interfaces in platforms like ChatGPT and Grok (Fawns et al., 2022). For workload reduction, experienced teachers reported greater benefits (M = 4.1) than their counterparts (M = 2.9), with a significant difference of t (118) = 5.13, p < 0.001, indicating that generative AI is seen as a valuable time-saving tool in preparing lessons, automating feedback and managing grading (Woolf et al., 2013). Hence, perceptions regarding tailoring content showed the most substantial divergence (t (118) = 6.02, p < 0.001), reflecting that AI-equipped teachers acknowledge its superior capability to personalise materials, align instruction to learning styles and support differentiated learning (Luckin et al., 2016). The results are depicted in Table 2.
Impact of generative AI on personalised learning
The study revealed a significant positive impact of generative AI tools on implementing personalised learning strategies in Tanzanian secondary schools. Teachers reported that using AI-supported tools facilitated customised content delivery, increased student engagement and improved academic performance, particularly among low-performing learners.
Examples of AI-customised materials
Teachers used generative AI (e.g. ChatGPT, Curipod, or MagicSchool.ai) to create automatically generated quizzes that included adaptive difficulty settings and content mapped to national syllabus objectives, and to generate weekly lesson plans aligned with student diagnostic profiles. The results indicate statistically significant gains in engagement, participation and performance, supporting prior global findings (Holmes et al., 2021; Zawacki-Richter et al., 2019), as shown in Table 3.
Discussion of the results
The demographic distribution highlighted the representativeness and generalisability of the findings within Tanzanian educational contexts. Also, the study showed how promising generative AI can be as a practical partner in teaching, especially in Tanzanian secondary schools. The diverse group of 120 teachers from urban and rural schools shows a well-balanced teaching population regarding age, experience and qualifications. This range allowed the study to capture perspectives representing fundamental classroom dynamics across different regions and resource settings.
One of the most encouraging findings was the significant improvement in learning outcomes when teachers used AI tools to personalise instruction. Compared to traditional teaching methods, the AI-assisted approach scored notably higher in key areas like student engagement, the ability to meet individual learning needs, and the speed and quality of feedback, as supported by Baidoo-Anu and Ansah (2023), Marques-Cobeta (2024) and Octavio et al. (2024). These changes were not just minor improvements; student engagement jumped from 2.8 to 4.3 on a 5-point scale, while average test scores rose from 61% to 75%. Teachers also reported that students became more curious, involved and independent when interacting with AI-generated content like quizzes and scenario-based activities. These findings echo earlier international studies that link personalised learning with better academic performance and motivation (Holmes & Porayska-Pomsta, 2023; Zhao et al., 2024).
From the teachers’ perspective, the benefits extended beyond just student outcomes. Many educators noted that AI tools significantly cut down the time spent on lesson preparation and grading, addressing a common barrier to personalised, student-centred teaching. With generative AI managing routine tasks, teachers could dedicate more time to individual student support and innovative teaching strategies. This reflects one of the major challenges in Tanzanian classrooms: large class sizes combined with limited planning time. Instead of replacing teachers, AI tools functioned as dependable assistants, enhancing efficiency and responsiveness to student needs. Statistical analysis corroborated these views. Teachers experienced with generative AI consistently found the tools more user-friendly, effective in reducing their workload, and better at tailoring content to student requirements. The disparity in their feedback compared to teachers lacking AI experience was statistically significant (p < 0.001 across all metrics), indicating that increased use of these tools led to greater appreciation among teachers.
However, some teachers did not feel completely prepared. While most respondents were positive, many – particularly those in rural areas – lacked the necessary training and exposure to the technology. This reveals a significant gap: for AI to genuinely transform education, schools and policymakers need to invest in professional development and infrastructure that assists teachers at all levels of experience.
Overall, the findings indicate that generative AI can significantly impact classrooms, not by replacing teachers, but by equipping them with the tools they need to excel in teaching, adapting and inspiring. As Tanzania updates its education system, this study provides compelling evidence that generative AI, when thoughtfully applied, can help create a more inclusive, engaging and effective learning environment.
The strong consensus among teachers regarding generative AI’s ability to customise lessons supports the idea that it effectively enhances personalisation, aligning perfectly with educational theories that highlight learner-centred approaches (Tomlinson, 2017). The increase in student engagement reported by teachers emphasises the pedagogical benefits of tailored content, reinforcing Vygotsky’s theory regarding effective scaffolding that meets individual learner needs.
The high level of teacher satisfaction further establishes generative AI as a supportive partner, rather than a substitute, mirroring global research that encourages collaboration between humans and generative AI (Baidoo-Anu & Ansah, 2023; Vygotsky, 1987; Zhao et al., 2024). The additional findings underscore the real benefits of AI on students’ academic performance, offering quantitative data to back qualitative observations (Zhang & Zheng, 2020). Nevertheless, the challenges identified spotlight gaps in infrastructure and training, particularly pronounced in rural schools, highlighting the critical need for targeted investments in digital literacy, infrastructure and continuous professional development, as depicted by Brown (2003) and Shahzada et al. (2021). These insights reinforce the transformative potential of generative AI to enhance educational outcomes, improve teacher effectiveness and promote student achievement across Tanzania’s varied educational landscapes, strongly advocating for broader policy implementation.
Conclusion and recommendations
This study highlights the significant potential of generative AI to enhance personalised learning in Tanzanian classrooms by empowering teachers and improving student outcomes. There were limitations, including the fact that most teachers had insufficient training in TPACK, which was a significant barrier, especially in rural areas. Also, the affordability of the AI model is not easy. With observable benefits in engagement, performance and a reduced workload, AI can serve as a supportive partner for educators rather than a replacement. To fully harness this potential, we encourage national investment in digital infrastructure, targeted teacher training and thoughtful integration of AI within the curriculum. Education stakeholders are urged to view generative AI as a powerful tool for fostering inclusive and scalable transformation. When applied effectively, these technologies can play a vital role in addressing educational disparities and equipping students for a future driven by innovation and the demands of a knowledge-based economy.
References
| Alam, A. (2023). Intelligence unleashed: an argument for AI-enabled learning ecologies with real world examples of today and a peek into the future. AIP Conference Proceedings, 2717(1), 030001-1–030001-9. https://doi.org/10.1063/5.0129803 |
| Aleven, V. et al. (2023). Domain modeling for AIED systems with connections to modeling student knowledge: a review. In B. du Boulay (Ed.), Handbook of Artificial Intelligence in Education (pp. 127–169). Edward Elgar Publishing. |
| Arias-Flores, H. et al. (2025). Artificial intelligence and assistive technologies: a systematic review of educational applications for disabilities. In M. Antona & C. Stephanidis (Eds.), Universal Access in Human-Computer Interaction (pp. 283–292). Springer Nature Switzerland. |
| Baidoo-Anu, D., & Ansah, L. O. (2023). Education in the era of generative artificial intelligence (AI): understanding the potential benefits of ChatGPT in promoting teaching and learning. Journal of AI, 7(1), 52–62. https://doi.org/10.61969/jai.1337500 |
| Bernacki, M. L., Greene, M. J., & Lobczowski, N. G. (2021). A systematic review of research on personalized learning: personalized by whom, to what, how, and for what purpose (s)?. In B. du Boulay (ed.), Educational Psychology Review, 33(4), 1675–1715. https://doi.org/10.1007/s10648-021-09615-8 |
| Bigham, J. P., Lin, I., & Savage, S. (2017). The effects of ‘not knowing what you don’t know’ on web accessibility for blind web users. In: Proceedings of the 19th International ACM SIGACCESS Conference on Computers and Accessibility (pp. 101–109). |
| Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2), 77–101. https://doi.org/10.1191/1478088706qp063oa |
| Brown, T. H. (2003). The role of m-learning in the future of e-learning in Africa. 21st ICDE World Conference, 110, 122–137. |
| Chen, X. et al. (2021). Past, present, and future of smart learning: a topic-based bibliometric analysis. International Journal of Educational Technology in Higher Education, 18(1), 2. https://doi.org/10.1186/s41239-020-00239-6 |
| Creswell, J. W., & Clark, V. L. P. (2017). Designing and Conducting Mixed Methods Research. Sage Publications. |
| Dwivedi, R. et al. (2023). Explainable AI (XAI): core ideas, techniques, and solutions. ACM Computing Surveys, 55(9), 1–33. https://doi.org/10.1145/3561048 |
| Filgueiras, L. V. L., Prietch, S. S., & Preti, J. P. D. (2015). Empowerment of assistive technologies with mobile devices in a DUI ecosystem. Procedia Computer Science, 67. 358–365. https://doi.org/10.1016/j.procs.2015.09.280 |
| Holmes, W. (2020). Artificial intelligence in education. In A. Tatnall (Ed.), Encyclopedia of Education and Information Technologies (pp. 88–103). Cham: Springer International Publishing.Switzerland. |
| Holmes, W. (2023). In W. Holmes & K. Porayska-Pomsta (Eds.), The Unintended Consequences of Artificial Intelligence and Education, Routledge. |
| Holmes, W., & Porayska-Pomsta, K. (2023). The Ethics of Artificial Intelligence in Education. Routledge. |
| Holmes, W. et al. (2022). Ethics of AI in education: towards a community-wide framework. International Journal of Artificial Intelligence in Education, 32(3), 1–23. https://doi.org/10.1007/s40593-021-00239-1 |
| Huang, X. et al. (2023). Trends, research issues and applications of artificial intelligence in language education. Educational Technology & Society, 26(1), 112–131. |
| Ishengoma, F., & John, E. (2024). Factors influencing the adoption of mobile-based AI services in Tanzanian manufacturing SMEs. Vilakshan-XIMB Journal of Management, 22(2), 234–252. https://doi.org/10.1108/XJM-11-2023-0214 |
| Joyce-Gibbons, A. et al. (2018). Mobile phone use in two secondary schools in Tanzania. Education and Information Technologies, 23(1), 73–92. https://doi.org/10.1007/s10639-017-9586-1 |
| Kasneci, E. et al. (2023). ChatGPT for good? On opportunities and challenges of large language models for education. Learning and Individual Differences, 103, 102274. https://doi.org/10.1016/j.lindif.2023.102274 |
| Kitalima, T. (2024). Teachers’ knowledge of the use of communication strategies: a case of kiswahili as a second language classroom. Kiswahili, 87(1), 132–146. |
| Kiunsi, R. (2013). The constraints on climate change adaptation in a city with a large development deficit: the case of Dar es Salaam. Environment and Urbanization, 25(2), 321–337. https://doi.org/10.1177/0956247813489617 |
| Laak, K.-J., & Aru, J. (2024). AI and personalized learning: bridging the gap with modern educational goals. ArXiv preprint p1-9 ArXiv, abs/2404.02798. https://doi.org/10.48550/arXiv.2404.02798 |
| Lin, L. et al. (2024). The personalized learning by interest effect on interest, cognitive load, retention, and transfer: a meta-analysis. Educational Psychology Review, 36(3), 88. https://doi.org/10.1007/s10648-024-09933-7 |
| Maritim, E. K., & Mushi, H. M. K. (2012). Mobile technologies for enhancing distance learning in Tanzania: an exploratory study Ezra. Huria Journal of the Open University of Tanzania, 13(Special Issue), 123–138. Retrieved from https://www.out.ac.tz/page.php?m=179 |
| Marques-Cobeta, N. (2024). Artificial intelligence in education: unveiling opportunities and challenges. In: Innovation and Technologies for the Digital Transformation of Education: European and Latin American Perspectives (pp. 33–42). |
| Miao, F., & Holmes, W. (2021). Artificial Intelligence and Education. Guidance for Policy-Makers, United Nations Educational, Scientific and Cultural Organization (UNESCO): Paris, France. |
| Mnyawami, Y. N., Maziku, H. H., & Mushi, J. C. (2022). Enhanced model for predicting student dropouts in developing countries using automated machine learning approach: a case of Tanzanian’s Secondary Schools. Applied Artificial Intelligence, 36(1), 2071406. https://doi.org/10.1080/08839514.2022.2071406 |
| Molenaar, I. (2022). Towards hybrid human-AI learning technologies. European Journal of Education, 57(4), 632–645. https://doi.org/10.1111/ejed.12527 |
| Mosavi, A., Ozturk, P., & Chau, K. (2018). Flood prediction using machine learning models: literature review. Water, 10(11), 1536. https://doi.org/10.3390/w10111536 |
| Mramba, N. R. (2024). The potentials of artificial intelligence in improving Africa informal cross border trade. What works, what doesn’t, and what’s next to Africans? African Journal of Land Policy and Geospatial Sciences, 7(1), 92–112. |
| Mtebe, J. S., & Raphael, C. (2018). Key factors in learners’ satisfaction with the e-learning system at the University of Dar es Salaam, Tanzania. Australasian Journal of Educational Technology, 34(4), 107–122. https://doi.org/10.14742/ajet.2993 |
| Nzeng’e, D. M., Gathogo, N., & Kamunyu, R. (2021). Effect of internal team environment on church growth in Pentecostal Churches in Kenya. The International Journal of Humanities & Social Studies, 9(6), 158–167. https://doi.org/10.24940/theijhss/2021/v9/i6/HS2106-050 |
| Octavio, M. M., Argüello, M. V. G., & Pujolà, J.-T. (2024). ChatGPT as an AI L2 teaching support: a case study of an EFL teacher. Technology in Language Teaching & Learning, 6(1), 1142. https://doi.org/10.29140/tltl.v6n1.1142 |
| Ponera, J. M., & Madila, S. S. (2024). Harnessing the Use of Artificial Intelligence among Higher Education Institutions in Tanzania: Challenges and Prospects, Edukasiana: Jurnal Inovasi Pendidikan, 3(3), 269–279. https://doi.org/10.56916/ejip.v3i3.714 |
| Ranganai, N. et al. (2022). Challenges and opportunities for digital inclusion in marginalised communities. In: Digital Transformation for Promoting Inclusiveness in Marginalized Communities (pp. 72–94). |
| Said, S. R. (2018). ICT Accessibility Solutions to Persons with Visual Impairment at the Open University of Tanzania. The Open University of Tanzania. |
| Sedoyeka, E. (2012). Obstacles in bridging the digital divide in Tanzania. International Journal of Computing & ICT Research, 6(1), 64–75. |
| Selwyn, N. (2021). Education and Technology: Key Issues and Debates. Bloomsbury Publishing. |
| Shahzada, G. et al. (2021). Are students of secondary schools of seven districts different on their self-estimates of multiple intelligences? A case study of southern districts of Khyber Pakhtunkhwa. Frontiers in Education, 6, 679289. https://doi.org/10.3389/feduc.2021.679289 |
| Tellman, B. et al. (2021). Satellite imaging reveals increased proportion of population exposed to floods. Nature, 596(7870), 80–86. https://doi.org/10.1038/s41586-021-03695-w |
| Thomas, J. (2023). Evaluation of Personalized Learning, [Doctoral dissertation, University of Texas at Tyler]. UT Tyler Scholar Works. http://hdl.handle.net/10950/4262. |
| Tomlinson, S. M. (2016). Perceptions of accessibility and usability by blind or visually impaired persons: a pilot study. Proceedings of the Association for Information Science and Technology, 53(1), 1–4. https://doi.org/10.1002/pra2.2016.14505301120 |
| UNESCO. (2022). Recommendation on the Ethics of Artificial Intelligence. Retrieved from www.unesco.org/open- |
| Vygotsky, L. S. (1987). The Collected Works of LS Vygotsky: The Fundamentals of Defectology (Vol. 2). Springer Science & Business Media. |
| Wang, H. et al. (2025). Exploring the relationship between teachers’ perceived workload, challenge-hindrance stress, and work engagement: a person-centered approach. BMC Psychology, 13(1), 1–19. https://doi.org/10.1186/s40359-025-02537-y |
| Xia, Q. et al. (2023). The mediating effects of needs satisfaction on the relationships between prior knowledge and self-regulated learning through artificial intelligence Chabot. British Journal of Educational Technology, 54(4), 967–986. https://doi.org/10.1111/bjet.13305 |
| Zhang, J., & Zheng, X. (2020). The influence of schools’ organizational environment on teacher collaborative learning: a survey of Shanghai teachers. Chinese Education & Society, 53(5–6), 300–317. https://doi.org/10.1080/10611932.2021.1879553 |
| Zhao, J., Chapman, E., & Sabet, P. G. P. (2024). Generative AI and educational assessments: a systematic review. Education Research and Perspectives (Online), 51, 124–155. https://doi.org/10.70953/ERPv51.2412006 |