Students tackle realistic scenarios through interactive simulations. These digital environments adapt to student choices, requiring them to use AI tools to navigate complex, real-world data problems.
back
Israel is redesigning national exams to reflect how professionals actually work. A pilot in Data Analysis replaces static, memorization-based tests with AI-integrated simulations — assessing how students direct AI, validate outputs, and reason through real-world problems. The goal: graduates who can think with AI, not just pass a test.
subjects:
pedagogy
The Challenge
One of the key challenges facing education systems today is the gap between traditional high-stakes examinations and the skills required in modern professional life. While many exams still emphasize memorization and individual recall, contemporary workplaces increasingly require data literacy, critical thinking, problem-solving, and the ability to work effectively with digital and AI-based tools.
This challenge calls for a broader modernization of national assessment: moving from static exams toward authentic tasks that measure higher-order thinking, process quality, and the responsible use of AI.
The Solution
As a first step toward a more significant transformation of national evaluation, a pilot has been launched within the field of Data Analysis. This project serves as a practical laboratory to test and refine methodologies that integrate Artificial Intelligence across all stages of the assessment process. By bridging the gap between secondary education and the digital economy, the initiative ensures that standards remain relevant to the modern workforce. This approach moves beyond using AI as a simple tool, establishing a workflow that mirrors professional environments and allows students to tackle complex problems previously inaccessible in traditional formats.
The assessment is organized around three stages that mirror professional analytical work:
- Investigation: defining real-world problems and searching for relevant data.
- Analysis: exploring datasets to uncover patterns using databases, visualization platforms, and AI.
- assistants Strategic Synthesis: developing reasoned recommendations, with human judgment at the center.
The Framework
The model evaluates student agency and analytical depth. The pilot uses a multidimensional assessment structure built on two pillars:
- Competency-Based Framework: Priority is given to data literacy, critical thinking, and advanced prompt engineering.
- Process-Oriented Grading: Evaluation is based on the quality of prompts, the validation of AI outputs, and the depth of iterative reasoning.
To implement these pillars and maintain high academic standards, the pilot uses a multi-layered structure that evaluates the entire problem-solving process:
-
AI-Driven Simulations
-
Prompt Engineering and Logic
Evaluation focuses on the quality of the dialogue between the student and the AI. Grades are awarded for the ability to write precise prompts, validate AI outputs, and iteratively refine results until a reliable solution is reached.
What This Framework Makes Possible
This pilot serves as a scalable roadmap for international education systems. It bridges the gap between academics and real-life modern technology and skills, through empowerment of the Agentic Learner, and a strategic structure for measuring and evaluating. The ultimate objective is to utilize the insights gained from this experiment to transform assessment methods across diverse academic disciplines, ensuring that human judgment and AI collaboration remain the keys to future success.