New School Guidelines in Japan Emphasize AI Education

2. Upholding Academic Integrity: Evolving Approaches in the Age of Generative AI

In response to the proliferation of AI writing tools and generators, the guidelines have strengthened their position on academic integrity. MEXT has established clear protocols for distinguishing between AI-assisted work and original student contributions. These protocols include requirements for proper attribution when AI tools are used and specific guidelines for acceptable AI use in academic work.

Schools like Osaka’s Sakuragaoka Academy now use platforms like Turnitin’s AI Detector to differentiate human and machine-generated work.

MEXT recognizes the challenge is not to simply ban AI, but to educate students on how to navigate this powerful technology responsibly in academic and professional settings. Discussions are ongoing about how to revise academic integrity policies and assessment strategies to effectively address generative AI.

The focus is shifting from simple prohibition to:

• Educating students about the ethical use of AI tools: This includes understanding appropriate citation, acknowledging AI assistance, and differentiating between using AI as a tool for learning versus submitting AI-generated work as solely their own.
• Developing assessment methods that are less susceptible to AI misuse: This means moving away from rote memorization and fact-based recall towards tasks that emphasize critical thinking, problem-solving, creativity, and application of knowledge – skills where human input remains essential.
• Exploring the potential of AI as a learning tool when used ethically: Some educators are beginning to explore how AI writing tools can be used for brainstorming, drafting, and receiving feedback as part of the learning process, under strict guidance and with clear ethical boundaries.

Did You Know? 67% of Japanese universities report AI-assisted plagiarism cases (Mainichi Shimbun, 2023).

3. Modernizing Assessment Methods: Beyond Rote Learning to 21st-Century Skills

In 2021, the Organisation for Economic Cooperation and Development (OECD) published a report underscoring the need to reimagine assessment in an age of intelligent machines and advocating for more open-ended, skills-based evaluations. MEXT’s guidelines mirror these proposals.

MEXT’s recommendation to revisit traditional assessments is becoming more concrete. Drawing inspiration from the OECD’s work and global best practices, the emphasis is increasingly on:

• Performance-based assessments: Projects, presentations, debates, and real-world problem-solving tasks that demonstrate deeper understanding and application of skills, rather than just memorization.
• Portfolio assessments: Collecting student work over time to showcase growth in competencies like computational thinking, creativity, collaboration, and communication.
• Formative assessment and feedback: Using ongoing assessment to guide learning and provide personalized feedback, rather than relying solely on high-stakes summative exams.
• Integrating computational thinking into assessments across subjects: This doesn’t mean just coding tests, but assessing the ability to break down problems, identify patterns, design algorithms, and think logically – skills applicable across disciplines.

While standardized testing still exists in Japan, there’s a growing recognition that assessments need to evolve to measure the skills needed for an AI-driven future. Pilot programs and research are underway to explore alternative assessment methods.

In Fukuoka, students design AI chatbots to solve community issues—a task assessed on creativity, collaboration, and problem-solving.

Traditional evaluation methods are being reconsidered in favor of assessments that emphasize:

• Critical thinking and problem-solving capabilities
• Ability to effectively collaborate with AI tools
• Creative application of knowledge
• Communication skills in both human and AI-integrated contexts