Al and school students – The Hindu

The impact of AI on school education in India is likely to widen existing gaps among students, particularly those from lower socioeconomic backgrounds. This is a critical consideration that is often neglected in discussions about AI expansion, especially in the children’s learning space.

For students with access to AI and the ability to ask the right questions, it can become a valuable tool, sometimes even surpassing the best teacher. For instance, if a child or teacher understands how sequencing subject, verb, and objects improves text cohesion, AI can quickly and accurately identify these components far more effectively than a human. This is also true for many other components that assist in selecting appropriate content for a given context.

In mathematics, AI can help students distinguish between questions that explain mathematical procedures and those that relate to relevant concepts. This is also true for science questions. Finally, in social sciences, AI can help students determine which information enhances their understanding or adds value to their audio or written text.

However, access to such tools and opportunities to reach this level of proficiency, which involves cost, is easier for students from private elite schools than those from government schools. Furthermore, within government schools, there are significant variations. For example, a student in a remote government school in Bihar will have a vastly different experience compared to a student in a government school in Mumbai.

When used intelligently and prudently, artificial intelligence has the potential to revolutionise school education in the country. It can enhance learning and serve as an extended academic resource for teachers, enabling them to ask questions at the appropriate level before they reach the learner’s level.

In retrospect, it is possible that such interventions may resemble the efforts made in the early 1990s of the last century to bring quality learning through pan-India projects like the Department of Primary Education Programme (DPEP) and its predecessor, the School System Improvement Act (SSA). Both these quality intervention projects became overly focused on teaching teachers how to teach, becoming so mechanical that trainers often communicated beyond the immediate comprehension and needs of teachers. For instance, university-trained resource persons were found interacting with teachers to determine whether language evolved or was innate to humans, or whether mathematics is concrete or abstract. While these discussions sometimes sparked interest in teachers, they ultimately did not compensate for their missing knowledge to improve classroom teaching. Terms like ‘concept learning’ were often so abstract to explain and equally difficult to comprehend. Resource persons sometimes suffered from the ‘Curse of Knowledge’ as they forgot the time they took to comprehend these ideas, which they wanted teachers to know and use within a few days of training. Moreover, teachers were left alone to build on what was transferred to them in the training without support to clarify their doubts or to read more about the knowledge they acquired, either as a question or a piece of information.

Artificial intelligence with chat capabilities could address these gaps, provided teachers learn to ask the right questions. Questions that help in the process of arriving at an answer and questions that explain the building blocks of knowledge. Asking a wrong question, even in ChatGPT, can lead to memorisation. This leaves us with the challenge of building teachers’ capacity to distinguish between procedure questions and questions that help locate the building blocks of knowledge. If AI is to deliver for those concerned with universal education to grow horizontally, this is a monumental task for school education.

prasoon68@icloud.com