ASEF Shanghai Manifesto
Thirteen principles of the Shanghai Manifesto on Universities’ Role in Developing Skills for an AI-Powered Future ("Artificial Intelligence", AI) have been developed as a call to inquiry and a call to action. They reflect the shared conviction of the Shanghai Cohort of the 6th ASEF Higher Education Innovation Laboratory ("ASEFInnoLab6", October 2025) that AI represents not merely a technological shift but a societal and ethical turning point that requires universities to rethink their missions, pedagogies and partnerships for the common good.
Principles
1. Put People Before Algorithms. AI must enhance, not eclipse, human purpose, creativity, and agency.
Example: In university admissions or recruitment, AI can assist by screening candidates, but the final decision must always rest with a human. Clear appeal mechanisms should exist to challenge algorithmic outcomes, ensuring fairness and human oversight remain central to every evaluative process.
2. Maintain the Right to Think Differently. Protect diversity of thought and culture against the homogenising power of machine logic.
Example: Academic communities should safeguard the freedom to choose how AI is integrated into their work. For instance, institutions should guarantee opt-in and opt-out options for AI-supported teaching and learning or research tools, encouraging diversity of academic practices and thought.
3. Make Ethics Non-Negotiable. Progress requires a moral compass, otherwise any innovation will remain fruitless in the long-term.
Example: Every AI-related project, course, or research activity should undergo an ethical review process. Universities should establish or strengthen ethical guidelines that govern AI’s use in research, teaching, and learning to ensure technology development remains morally sound.
4. Bridge the AI Divide. More equitable access to AI tools, skills, and opportunities is a shared global responsibility.
Example: Universities should form partnerships with industry, civil society and governments to provide equitable access to AI education and tools. Initiatives such as shared infrastructure, capacity development, training programmes, and mentorships can help underrepresented communities benefit from the AI revolution.
5. Tear Down the Walls of the Classroom. Learning should be lifelong, life-wide, and borderless.
Example: Universities should promote flexible learning opportunities such as MOOCs, Open Educational Resources (OER), micro-credentials, and community-based learning hubs. These formats make education accessible to learners of all ages and backgrounds, supporting continuous and inclusive skill development.
6. Turn Campuses into Living Labs. Universities must become open spaces for experimentation, co-creation, and civic innovation.
Example: Universities can serve as real-world laboratories by encouraging students, researchers, and citizens to co-create and test AI-driven solutions, campuses become active contributors to local and global progress.
7. Redefine Teaching and Learning. Rethink assessment and credentialing to value adaptability, reflection, inclusivity and human-AI collaboration.
Example: Emerging initiatives such as European university alliances and inter-institutional micro-credential systems demonstrate how AI can reshape learning pathways. These models promote flexibility, interdisciplinary collaboration, and recognition of diverse learning achievements across borders.
8. Enforce Regulations Based on Evidence and Involve All the relevant Stakeholders. Governance of AI in education must include students, staff, private sector/businesses, and citizens alike.
Example: AI governance frameworks in education should be co-created by diverse learners with and without disabilities, faculty members, policymakers, and industry. Transparent, data-informed decision-making ensures regulations evolve with technology while protecting the interests of all members of the learning community.
9. Break the Disciplinary Silos. The challenges of AI demand cross-disciplinary, cross-sectoral, and cross-cultural learning. Example: Universities should incentivise interdisciplinary projects that unite computer scientists, social scientists, ethicists, and artists. Research grants or joint programmes can encourage innovative, sustainable, and holistic approaches to understanding and using AI.
10. Educate for Agency, Not Just Employability. Graduates must not only fit into systems—they must be equipped to question and reshape them.
Example: Beyond preparing students for the job market, universities should help them understand and question the systems they will work within. By embedding AI literacy, ethics, and critical thinking into curricula, graduates gain the power to shape a more equitable technological future.
11. Teach AI to Care About the Planet. AI must serve ecological intelligence and sustainability, not unchecked consumption.
Example: Universities can lead by creating AI tools that prioritise sustainability—such as environmental monitoring bots or green data systems. Courses can teach “eco-prompting” practices and host student challenges focused on designing environmentally conscious AI applications.
12. Demand Proof, Not Promises. Institutional AI strategies should be evidence-based, ethical, transparent, and accountable.
Example: Institutions should back every AI initiative with transparent data on effectiveness and impact. Dedicated quality assurance units for AI, participatory evaluation processes, and research-backed policies ensure accountability and continuous improvement.
13. Share and Co-Create Knowledge. Asia and Europe can lead by example—through openness, reciprocity, and solidarity in AI education.
Example: Asia and Europe can strengthen collaboration through open-access platforms, shared databases, and joint MOOCs. By exchanging research, best practices, and learning resources freely, both regions can model equitable and sustainable AI-driven education.
More information:
- Shanghai Manifesto on Universities’ Role in Developing Skills for an AI-Powered Future
- ASEFInnoLab6
- Asia-Europe Foundation
Source: ASEF