{"id":25581,"date":"2026-05-11T14:50:14","date_gmt":"2026-05-11T18:50:14","guid":{"rendered":"https:\/\/design.ncsu.edu\/graphic-design\/?p=25581"},"modified":"2026-05-11T14:50:16","modified_gmt":"2026-05-11T18:50:16","slug":"designing-the-future-of-intelligence-mgxd-students-take-on-ai-assisted-reporting","status":"publish","type":"post","link":"https:\/\/design.ncsu.edu\/graphic-design\/2026\/05\/11\/designing-the-future-of-intelligence-mgxd-students-take-on-ai-assisted-reporting\/","title":{"rendered":"Designing the Future of Intelligence: MGXD Students Take on AI-Assisted Reporting"},"content":{"rendered":"\n\n\n\n\n

Students in NC State’s Master of Graphic & Experience Design (MGXD) program recently collaborated with the NC State Laboratory for Analytic Sciences (LAS)<\/a> on a genuinely novel design challenge: how do you build an interface that helps intelligence analysts work with<\/em> AI to move sensitive information safely and efficiently through a tiered reporting system? Tiered reporting \u2014 the structured process of tailoring intelligence for different audiences based on their security clearance, expertise, and need-to-know \u2014 is critical work. It’s also painstaking, high-stakes, and ripe for thoughtful design innovation.<\/p>\n\n\n\n

Led by Professor Helen Armstrong, student teams began by going deep on the existing workflow, mapping pain points and friction across the reporting process. From there, they developed original UX\/UI concepts that integrate AI not as a replacement for human judgment, but as a collaborative partner \u2014 one that helps analysts sanitize content, calibrate detail levels, and deliver the right information to the right people without compromising security.<\/p>\n\n\n\n

The Design Question<\/h2>\n\n\n\n

The central design question driving the project: How might an interface automate the tiered reporting process so that analysts can more efficiently and knowledgeably team with AI?<\/em><\/p>\n\n\n\n

That question opened up three rich areas of exploration:<\/strong><\/em><\/p>\n\n\n\n

Oversight<\/strong> \u2014 How should an interface surface AI recommendations and flag data modifications so analysts always stay informed and in control as content moves across classification tiers?<\/p>\n\n\n\n

Agency<\/strong> \u2014 How do you design for human involvement in a more automated system, keeping analysts active, deliberative participants rather than passive observers?<\/p>\n\n\n\n

Trust Calibration<\/strong> \u2014 How can an interface honestly communicate what AI can and can’t do \u2014 especially for analysts at different experience levels \u2014 so that trust in the system is earned, accurate, and appropriately bounded?<\/p>\n\n\n\n

Student Prototypes<\/h2>\n\n\n\n
\n