Co-design and validate context-aware AI through DBIR blended with user-centered design in privacy-preserving enclaves¶

Strategy · All strategies
Evidence · no claims cited

Description¶

The article's application, testing, and validation strategy is to bring stakeholders together to co-design the LC and implement it in AI instructional tools using a design-based implementation research approach combined with user-centered design. "we propose a design-based implementation research (DBIR) framework blended with user-centered design (UCD)." Testing occurs by integrating LC-based AI tools into the OpenStax Assignable platform and studying engagement and outcomes via SafeInsights, with students and instructors engaged as active partners, particularly on ownership, hosting, and the privacy–informativeness tradeoff.

Design Implications¶

Context¶

Requirements¶

• Engage a diverse group of stakeholders—particularly students and instructors—as active partners in a User-Centered Design process

Constraints¶

• All research and pedagogical interventions are conducted within SafeInsights' privacy-preserving data enclaves; the LC presents significant privacy risks that must be addressed intentionally

Target Learners¶

• K-12 and college learners reached through the OpenStax platform's national reach

Target Learning Goals¶

• Validating that context-aware AI supports deep learning and reduces equity gaps

Related Strategies¶

• Secure enclaves for privacy-protecting researcher access to student-level data
• Involve key stakeholders directly in defining privacy and designing LA systems
• Co-create contextualized AI literacy policies via multi-stakeholder task forces with phased roadmaps

Examples¶

-

Key Sources¶

• Naiming Liu, Brittany Bradford, Johaun Hatchett, Gabriel Diaz, Lorenzo Luzi, Zichao Wang, Debshila Basu Mallick, Richard Baraniuk. (2025). Learning Context: A Unified Framework and Roadmap for Context-Aware AI in Education. https://arxiv.org/abs/2512.24362