A Transparency Index Framework for AI in Education

Transparency must be built into every stage of AI development for education, not just added at the end—the framework shows that transparency considerations need to be embedded "from the data collection stage until the AI system is deployed in the real world and iteratively improved," meaning educators and institutions should ask questions about AI systems from their earliest design phases.

Different stakeholders in education need different types of transparency information—the research involved educators, ed-tech experts, and AI practitioners and found that their transparency requirements vary, suggesting that one-size-fits-all explanations of how AI works in schools won't adequately serve all the people who need to understand and oversee these systems.

Transparency is foundational to other important ethical safeguards in educational AI—the paper demonstrates that transparency "enables the implementation of other ethical AI dimensions in Education like interpretability, accountability, and safety," meaning that without understanding how an AI system works, it becomes nearly impossible to ensure it's fair, safe, or that someone can be held responsible when problems occur.

The education sector has been overlooked in AI ethics discussions—while "numerous AI ethics checklists and frameworks have been proposed," the authors note that "no such work has been done on developing transparent AI systems for real-world educational scenarios," highlighting a gap where schools may be adopting AI systems without adequate transparency guidance tailored to their specific needs.