Ideas & Debate

Can Artificial Intelligence disrupt education?


AI will do things that no human being can do to maximise learning outcomes. FILE PHOTO | NMG

Until the advent of universal literacy some 150–200 years ago, education was mostly for priests, bureaucrats and specialists. Over time, public education has served to bring people closer to realizing their potential as human beings and become productive in their respective economies. In Africa, even with widespread public education, universal education is still a luxury for millions of people. This fact has forced Africa to lag behind other economies, as evidenced by the many developing countries on the continent.

With the emergence of fourth industrial revolution technologies such as Artificial Intelligence (AI), Africa can leapfrog development.

AI will change what we learn and how we teach and learn it, in the process help us to discover individual capabilities, enabling accessible and inclusive education with better utilisation of educational resources.

In recent times, psychologists have established that not every student learns by listening to teachers. Each student has particular abilities and needs. This makes it difficult for a teacher to adjust learning based on particular needs. AI, however, enables differentiation that is difficult for teachers especially if the class is large.

Several digital platforms have developed intelligent instruction leveraging AI to provide learning, testing and feedback to students from early childhood education to college that gives each student the challenges they are ready for.

They can also identify gaps in knowledge for each student and convey information on new areas of challenge for students to undertake. More importantly, AI system can read into the student’s expressions while undertaking a task and tell if a student is struggling.

The system will then adjust the lesson to respond to the student needs, learn the speed at which the student is able to comprehend and adjust accordingly, creating a personalised or customised learning system for each student. This does not mean that AI will replace teachers.

There is much more that happens in class than just teaching. Human beings still have an advantage over AI systems. They have soft skills such as empathy, a critical component for learning.

Teachers therefore will have more work either by providing the information that teaches AI systems – often referred to as machine learning – for the systems to be able to execute as learning assistants. AI systems work well with the assistance of teachers.

The learning assistant tool can be a smartphone, laptop, tablet or any computer coupled with good connectivity. In Kenya, we could very easily offer affordable education if the ongoing Digital Literacy Programme (DLP) covers the entire country. At the moment, 97 percent of the schools have broadband. A small incentive to acquire a simple tablet and some training either by teachers or volunteers on how to access the learning, we could leapfrog in education and perhaps create an educational revolution in Africa.

AI will do things that no human being can do to maximise learning outcomes. For example, the government’s admission into university programs based on a one-time performance in different subject clusters is in most cases flawed.

Many students end up doing courses that either they cannot cope with or something they may never want to do in their lives.

AI can solve this problem by analysing student performance over a period of time to predict his/her future abilities to undertake the course. The problem of matching student performance and career path is more pronounced in Africa because of massification of higher education.

If the meagre educational resources are to be utilized effectively, then AI algorithms will be the most effective tool. With the help of teachers and other student data, it is possible to predict a career path for each student.

Whereas studies show that the characteristics of a good doctor include empathy, humanity, honesty, compassion, understanding, courage, commitment and competence we often focus on grades to select medical school students.

The result is that we sometimes have doctors who possess characteristics that do not fit the profile of a doctor. Like with human beings, AI can be vulnerable to manipulation of its brain (algorithm) to create bias. Its use therefore must be put under scrutiny to avoid flawed outcomes.

AI is set to disrupt education. It will bring greater inclusivity and accessibility to education but we must avoid the narrative that it will displace people out of their jobs. We need embrace its ethical use as a strategy to enable the continent leapfrog.