Can AI Transform Education in LMICs?
A blog by Amy Nye, a Frontier Tech Hub Helpdesk lead
For more information on the Helpdesk, click here
Education is a global cornerstone for societal progress that is being prioritised by local governments as well as in the UN’s 2030 Agenda for Sustainable Development. Yet, many low- and middle-income countries (LMICs) face significant challenges in achieving their goals when it comes to learning outcomes, as well as in providing high-quality teacher continuous professional development (TCPD). The Frontier Tech Hub Helpdesk conducted a rapid desk review to explore how AI technologies could help support teachers and school leaders, leading to better learning outcomes and foundational skills in Sierra Leone and Tanzania, as well as in other LMICs.
Image credit: DALL’E - generated on 13.11.’24
The Education Challenge in LMICs
Sierra Leone and Tanzania, like many LMICs, grapple with systemic barriers to effective teaching. Issues include limited resources, poor infrastructure, insufficient training opportunities, and coordination gaps within the education system. For instance, only 8% of third graders in Sierra Leone can read simple text, while Tanzania faces an average student-to-teacher ratio in classes of 51:1.
The challenges faced by teachers are vast. AI tools hold a lot of potential in addressing those problems and reducing burdens on teachers, but they also pose significant operational problems in the context of LMICs.
How AI Can Support Teachers
There are many different AI tools that are currently available that can support teachers and school leaders. Most people are most familiar with generic AI tools, like ChatGPT (OpenAI), Copilot (Microsoft), Gemini (Google), and Meta AI (Meta), which can generate educational materials that fit the specific needs and context of the teacher. For example, teachers could develop materials like rubrics, personalized content for students, and lesson plans and tailor them to specific subjects and grade levels. Although these tools are somewhat generic and broad, they are very useful in education. Having only one tool that can perform multiple functions can be preferable for teachers compared to separate tools that can be unpredictable and have variable performance and continuity.
More tailored, specialized tools play a variety of roles, including supporting in lesson planning, providing real-time feedback, serving as a teacher coach or teacher assistant, and creating personalized teaching materials. These tools can streamline administrative tasks, enhance teacher coaching, and improve lesson delivery. The full report from the FT Hub Helpdesk explores the capabilities of 12 different tools, including TheTeacher.AI, Khanmigo, Edthena, and others. Although AI tools being applied to support teachers in LMICs are relatively limited, the conceptualization of how these tools might be applied has a broad reach. The report includes an analysis of where there are existing AI tools for supporting teachers and school leaders and where there are gaps. The authors map teacher tasks that could be supported by AI (e.g., generate lesson plans and content) against four use cases for AI tools: Document Summariser, Evidence Checker, Strategy Checker, and Coach.
Considerations for Implementing Responsible Tech
While the promise of AI in education is significant, implementing these technologies in LMICs requires careful planning and ethical considerations:
Cultural Contextualization: AI solutions must be tailored to local languages and cultural norms to ensure relevance and inclusivity.
Bias and Fairness: Ensuring AI systems are trained on diverse, representative data can prevent the reinforcement of existing inequalities.
Infrastructure Needs: Access to devices, reliable internet, and electricity remains a critical barrier to widespread adoption.
Teacher Training: Educators need support to effectively integrate AI tools into their workflows.
Beyond AI: Alternative and Complementary Solutions
AI is not a one-size-fits-all solution. There are many non-AI technology solutions that can accomplish the same objectives and that may be more feasible for implementation in the context of LMICs. In some cases, non-AI technologies, like solar-powered tablets or interactive voice response systems, may offer more practical and cost-effective ways to support teachers.
If you’d like to dig in further…
🚀Please visit the full report produced by the FT Hub Helpdesk
Date published: 7th February 2025
