News

Leveraging Agentic AI for smarter decision-making in education

Policy-makers need data to make informed decisions to improve students’ learning outcomes. Yet, learning assessment data, which are rich with insights on what and how students learn, are often underutilized due to their complexity and the resources and capabilities required for analysis.
Argentic AI

To tackle this issue, UNESCO partnered with Capgemini for the company’s 7th Global Data Science Challenge, a hackathon for teams across the world. This year’s challenge focused on harnessing advanced AI technologies, specifically Agentic AI, to develop an open source chatbot, turning complex data into actionable insights for education policy and practice.

The winning team, insAIghtED, was announced this week and presented an AI-driven ‘policy expert’ capable of analyzing over 30 million data points on 4th-grade children’s reading achievement and providing data-informed guidance to policy-makers, educators, parents, and students alike.

About the winning solution

The winning team built its model on the 2021 PIRLS dataset, a global survey administered by the International Association for the Evaluation of Educational Achievement. The dataset is invaluable for its rich, standardized information, enabling the identification of patterns and trends in children’s reading abilities across regions and education systems.

The team extended their analysis by integrating additional data sources, such as GDP, life expectancy, and even YouTube content, creating an AI powered data analysis tool that provides insights in a few minutes in accessible formats. 

Their solution combined advanced analytics with user-friendly outputs, including concise summaries, detailed explanations, and data visualizations, making even the most complex datasets understandable and actionable for education stakeholders.

Capgemini will now provide UNESCO with the source code and a knowledge transfer of the winning solution.

Tackling global learning challenges

“This innovation brings us closer to addressing the global learning crisis,” said Gwang Chol Chang, UNESCO’s Chief of Education Policy, who was part of the jury selecting the winning model. “It demonstrates how AI can bridge the gap between raw data and impactful education policy, paving the way for transformative change”.

Today, 251 million children are out of school and 70% of aged 10 children in low- and middle-income countries are unable to read and comprehend a simple text. 

The need for innovation in education policy and practice is undeniable. Understanding which learners are making progress, which are not, and why is critically important for education leaders and policy makers to prioritize the interventions and education policies effectively. 

UNESCO’s partnership with Capgemini’s GDSC 2024 underscores the transformative potential of AI in and for education. By combining human creativity with cutting-edge technology, it lays the foundation for a future where data-driven insights enable effective, equitable, and inclusive education for all.