Once considered science fiction, generative AI tools have quickly become a real-world companion in classrooms around the globe. With AI reshaping education systems worldwide, the Institute of Informatics and Development (IID) brought the context of Bangladesh into the global conversation. On March 24th, 2025, IID presented its study, Generative AI in Education: Current State, Future Directions, and Implications in Bangladesh, at the Comparative and International Education Society (CIES) Annual Conference. The research was featured in a panel exploring inclusion, innovation, and impact in the era of generative AI.
The study, presented by Maisha Tasnim and co-authored by Aruja Islam and Faria Rahman, set out to understand how this change is unfolding at home—what educators and learners in Bangladesh know about generative AI, how they feel, and how prepared they are for this emerging learning tool. It also examines both the benefits and risks of integrating generative AI into Bangladesh’s education system, with a particular focus on implications for policymaking.
To capture the full picture, the research employed a mixed-methods approach. Qualitative data were collected through in-depth interviews with 10 educators, while quantitative data were gathered from a survey involving 205 students and teachers. Using the Theory of Planned Behavior, the research looked beyond opinions to understand how attitudes, social norms, and perceived control influence adoption.
The study found that while about 8 out of 10 students claim a moderate understanding of AI, confidence in advanced skills—especially among younger students—is low. Only a small number of secondary-level students can use AI tools proficiently, revealing a widening digital skills gap. The gap in AI knowledge becomes even more evident when examined across different geographic areas. Students from City Corporation areas reported the highest levels of perceived AI knowledge (16.8%), followed by those from municipalities (11.1%) and villages (10.8%). In contrast, no students from hilly or coastal regions reported any familiarity with AI. Since the online survey was voluntary, participation was largely dominated by interested individuals, most of whom were from urban areas with better internet access. As a result, the actual disparity in AI knowledge—particularly among students in rural and remote regions—is likely even wider than the survey data suggests.
Beyond academic tasks, AI usage also extends into students’ personal lives, such as chatting with AI for quick answers, using it for writing emails, seeking entertainment, or asking health-related questions. Teachers, too, are tapping into AI—mostly for lesson planning, grammar checks, simplifying complex topics, and speeding up task completion. About 4 out of 5 educators reported using AI tools to save time and support learners who need extra help.
Many of them recognize AI’s benefits for student engagement and support, but their optimism is tempered by concerns over misuse and overreliance. Around 6 in 10 educators worry students are getting too dependent on AI, using it more as a shortcut than a study tool. That reliance, they warn, could weaken skills like critical thinking and creativity. Teachers themselves face uncertainty, having received little to no institutional training. Most have had to explore AI independently, often without clear guidance on ethical use or how to assess the accuracy of AI-generated content.
This lack of support extends to students as well. Schools remain inconsistent—some permit AI use, others ban it, and many have yet to take a clear stance. As a result, students are left to navigate this emerging technology on their own, mirroring the experience of their teachers. Both groups are eager to engage with AI but are doing so without the necessary direction from educational institutions.
As a result, both teachers and students across Bangladesh are calling for clear policies, structured learning frameworks, and dedicated training programs. Many want to see AI literacy added to the national curriculum—not just to teach students how to use these tools, but how to use them wisely. Educators also emphasized the need for a balanced approach to AI integration in education, highlighting the importance of dedicated training programs and clear data privacy policies.
Building on these calls for structured support, the broader findings of the study reveal that while AI is gradually becoming part of the educational experience, its integration remains fragmented and unequal. The lack of formal training and coherent policies leaves both students and educators navigating this shift largely on their own. Without inclusive, forward-thinking policies that address both training and access, the digital and AI skills gap risks becoming even more pronounced, deepening educational inequalities across the country.