Bespoke Learning: Using AI to Upscale Quality Education

Conventionally, development has always followed education. Americans took nearly a century to make primary education universal, and another 50 years to develop secondary education. Later, other nations followed this model - the Japanese after the Meiji restoration in 1868, Russia after WWI, and China after the communist revolution.

Even before Independence, India recognized the value of education. During his 1931 visit to Russia, Rabindranath Tagore observed that the success of Russian education lay “not merely in the numbers being educated, but the thoroughness, intensity of education”.

Rabindranath Tagore's insight emphasizes that high-quality education must be accessible to all for it to lead to national development.

The 2023 ASER report reveals the state of education in India, noting significant progress in school attendance (over 80%). However, the report paints a disheartening picture of education quality: only about 25% of 14-18-year-olds could read a standard II textbook in a regional language; slightly more than half could read sentences in English; 39% could measure distance using a scale if the starting point was not zero; and only 42% of Standard V students could read at or above a Standard II level.

The message is that there is an urgent need to improve education quality across the country to prevent the current generation of students from missing out on future job and income opportunities.

One major cause of low-quality education is the conventional schooling system's “one-size-fits-all” approach, which fails to account for individual learning speed and methods in the same classroom. This often results in slow learners falling behind, accumulating a learning deficit over time.

The challenge lies in identifying and addressing learning deficits as they occur, requiring teachers to assess students in real-time and tailor instruction to each child's unique needs.

Artificial Intelligence (AI) offers potential solutions to these challenges, as outlined below:

The Chatbox

A chatbox permits individual attention to be given to each student, at any time of day, without getting tired. It can adapt to local contextual conditions as well. Most importantly, chatboxes can be tailored to different learning styles and fine-tune what and how to teach, depending on the interaction with each child. For example, this could involve testing pupils’ understanding of ideas in real-time by asking them to answer questions. In other words, the chatbox adapts to classroom teaching in real time depending on the child’s progress.

ChatGPT

ChatGPT has the potential to empower teachers to regularly revise their class notes and master their curriculum - a key determinant of quality teaching. This results in below-par teachers and even modestly qualified teachers performing as well as their better-performing peers.  

A study of bridge schools in Kenya by Nobel Laureate-winning economist Michael Kremer and others highlights this: modestly qualified teachers were given “scripts” for lessons, delivered on tablets. On average, among the 10,000 students in these bridge schools, they found that in two years students had mastered nearly an extra year’s worth of the curricula, compared to pupils enrolled in conventional schools.

Digital technology buttressed by AI has the potential to empower teachers to deliver state-of-the-art individualized learning to students. For this, State Governments have to re-align their policies and repurpose State Council of Educational Research and Training (SCERT) and District Institute of Education & Training (DIET) to promote a kind of bespoke learning highlighted above. This will help bridge the quality gap in education for many students and open up huge opportunities in life.