By Dr Lalit Sharma
Artificial Intelligence in the education system is a much talked-about theme across the world. However, the realization of its potential is not uniform across developed and developing countries. The application of Artificial Intelligence (AI) algorithms and systems in education is growing by the year. As education evolves, researchers are applying advanced AI techniques, i.e. deep learning, data mining, etc., to deal with complex learning issues and customize teaching methods for individual students.
To add to the list, COVID-19 accelerated the process when lectures, classes, and test-taking were taken online. As revealed by a 2020 Educause poll, 54% of institutions were using online or AI-based remote proctoring services.
AI in education initially involved use of computers and computer-related systems. It has advanced further with the emergence of web-based and online education platforms. Embedded systems enable the use of robots as colleagues or independent instructors and chatbots can perform teacher-like functions.
Studies suggest that the use of these platforms and tools have improved teacher effectiveness, instructional quality, and learning experiences for students. It facilitates the customization of learning materials to the needs and capabilities of the learners. AI is being explored extensively in K-12 education, higher education, corporate training, language learning, quality training, and reading. However, there are several challenges in adopting AI for education.
Recent reviews suggest that research regarding AI in Education is mainly concentrated in developed countries and is limited in the developing world. There are multiple limitations or challenges that contemporary education faces in leveraging AI technology.
Challenge #1: An all-inclusive public policy on AI for development
Though AI holds incredible potential for improving education systems, the 360-degree inclusion of AI in education needs a robust policy support. Education leaders should be aided financially, as well as ethically, to focus on shaping learners who have the skills to thrive in the AI society. As of now, the AI advancements in schooling have been initiated by the private sector players such as UpGrad, Coursera, McGraw-Hill, Pearson, and IBM, while the government struggles to catch up.
The development of public policy for AI in education is in its earliest stages, but it’s expected to grow in the next 10 years. State policies should resolve inquiries to generate solutions and guidelines, support innovative ecosystems to realize the opportunities of AI in the field of education.
Several countries have already made significant budgetary commitments toward creating AI research centres, and recruiting and preparing AI experts. Governments are also investing in research and advanced training in AI by establishing academic centres of excellence in AI, scholarships, and research institute networks.
Countries like Argentina, Singapore, Estonia, Malaysia, Kenya, France, South Korea, and Germany have forged partnerships between industry and academia to share material and financial resources. Partnerships between universities and research institutes will foster collaborative research and intra-sector partnerships and advance academia-industry partnerships.
Challenge #2: Inclusion and Equity in AI in Education
AI may deepen the existing inequalities and divides because the disadvantaged populations might get excluded from AI-powered education because of the digital divide.
Equity and inclusion should become the core values when designing policies. The policymakers should consider these aspects while doing so:
• Urgency of creating a digital infrastructure in developing countries
• Learnings from previous experiences in terms of digital rights
• Close the educational gap between economically rich and poor students
• Close the gender gaps
Some of the obstacles to overcome include:
• ICT hardware unavailability
• Internet unreliability
• Data costs
• Lack of basic ICT skills
Multiple policies must be put in place to remove these basic obstacles. It is essential to consider internet as a human right and create multiple international alliances to build basic infrastructure in even the poorest sectors of the developing world.
Challenge #3: Preparing teachers for AI-powered education
Teachers remain at the frontline of education. They need to possess the knowhow on how AI-enabled systems can facilitate learning provision to make value judgments.
Training should emphasize on:
• Research and data analytical skills to interpret data provided by AI-enabled systems
• New management skills to manage human and AI resources at their disposal
• Critical perspective on how AI technologies affect human lives
• Take advantage of AI taking over repetitive tasks
• Help learners achieve new skills and competencies (that cannot be replaced by machines)
Challenge #4: Develop quality and inclusive data systems
The data we have is sporadic, unevenly distributed, and limited. A complete functional data analytics system can open possibilities for AI-enabled predictive and machine learning algorithms. However, the technologies covered for capturing data might prove costly for low- and middle-income countries. The costs need to be weighed carefully against the benefits.
Challenge #5: Ethics and Transparency
There are certain societal and ethical concerns to be addressed while implementing AI. Technology is improving quickly and what is impossible today can become possible tomorrow.
Data privacy and security is the immediate question that comes up in any discussion regarding data ethics. The challenge lies in using personal data while ensuring the protection of individual privacy preferences and personally identifiable information.
The collection and use of data must be anchored on express and informed consent, transparency, and fairness.
In conclusion, the private sector and tech giants are leading in most countries such as the United States and China. The rise of tech start-ups is also playing a significant role in accelerating AI adoption. Many countries have national AI strategies in which education is an element by default. In developing countries, the discussions and adoptions are limited. The adoption of AI in education should become a regular and continuous process. Only then can we hope to leverage the potential of technology to nurture today’s learners into tomorrow’s achievers.
The author is Associate Professor – IT and Business Analytics, Jaipuria School of Business, Ghaziabad
Disclaimer: Views expressed are personal and do not reflect the official position or policy of Financial Express Online. Reproducing this content without permission is prohibited.