Azra Hoosen | ah@radioislam.co.za
1 November 2024 | 17:30 CAT
3 min read

Artificial Intelligence (AI) is rapidly reshaping sectors across the globe, and education is no exception. As we explore the future of AI in this field, it’s essential to understand the potential benefits, ethical considerations, and the critical role of human oversight in its implementation.

In an interview with Radio Islam, Professor Nelishia Pillay from the University of Pretoria, a leading expert in AI and its applications for sustainable development, shared insights into the evolving role of AI in education. “If you look at education, and some of the things that have been done with trying to predict when someone could fail and then trying to put something in place to help them, if you look at intelligent tutoring systems that provide individual tuition the way a tutor would, all of those systems are developed based on human knowledge,” she explained.

These systems often rely on supervised learning, which involves using existing knowledge, such as symptoms, to diagnose a disease and inform AI decision-making.

However, Professor Pillay noted that inherent biases in human knowledge can sometimes filter into AI systems. This underscores the importance of human involvement in training and refining these systems. “When developing these systems, we need to keep in mind, in addition to assessing how accurate they are, we need to build in measures that assess how they can ensure that they don’t have biases, they don’t have ethical boundaries that are broken, etc.,” she said.

She also highlighted the need for governance specific to the industry and application of AI, adding, “There needs to be governance at different levels.”

In educational settings, for instance, ethical considerations are increasingly integrated into student submissions. Today, students are required to disclose any AI assistance used in their work, a practice that encourages responsible AI use in academic contexts.

“AI is just one emerging technology,” Professor Pillay observed, drawing parallels to earlier technological revolutions. Face-to-face education, she explained, has seen innovative applications of AI, such as online AI counsellors that connect students with real psychologists when necessary. Predictive tools in education, which can flag students at risk of academic failure, represent another exciting advancement in personalised learning. Yet, Professor Pillay believes in a “blended approach” to education, where AI supports but does not replace human-led learning.

Reflecting on her teaching experiences, she shared, “When you actually learn you go through a struggle process; you learn sometimes by getting things wrong.”

She emphasised that while AI can support the learning journey, it cannot replace the critical thinking and problem-solving process essential for mastering new skills. She acknowledged the differences across disciplines, suggesting that AI might play varying roles depending on the field of study.

She advised that proactive strategies, rather than reactive, are essential for managing the AI transition. She cited the example of banks transitioning online, which led to retrenchments, as a reminder of the need to stay ahead of technological shifts. AI’s increasing role in fields such as radiography—where its diagnostic accuracy sometimes surpasses human capabilities—demonstrates the potential for AI to transform industries, necessitating lifelong learning.

“We need to embrace lifelong learning; when you finish your degree, that’s not the end of it,” Professor Pillay advised as she advocated for continuous education.

She noted the critical role of tertiary institutions, industries, and possibly governments in fostering a culture of lifelong learning and adaptability. She explained that countries like Singapore, which offer annual stipends for citizens to upgrade their skills, provide a model for keeping the workforce agile in the face of rapid technological change.

According to Professor Pillay, “Those are the two solutions: lifelong learning as well as the infrastructure to facilitate that and secondly, to be at the forefront; use AI to work out what skills will be redundant to reduce the impact on unemployment.”

LISTEN to the full interview with Ml Junaid Kharsany and Professor Nelishia Pillay, from the University of Pretoria, here.