Self-directed portfolio sample · Branching scenario eLearning

Mission Possible: Responsible AI

A branching scenario experience with adaptive feedback and support based on learner decisions. Across three AI systems, learners investigate evidence, install safeguards, contain harm, and justify accountable responses. The experience is based on my From Risk to Responsibility white paper.

I created this self-directed sample for my portfolio to demonstrate my current approach to scenario-based eLearning and responsible AI.

I designed the concept, subject-matter approach, learning objectives, mission structure, scenarios, and decision points, refining the experience through repeated review and iteration. I used ChatGPT Codex to build the entire course, eliminating the need for a separate course-authoring tool or outside developer.

AI governanceBranching scenariosCodex
The experience

Governance decisions with visible consequences

01

Learning Path Review

Analyze AI-generated learning recommendations and determine what the evidence means for employee opportunity.

02

Rapid Course Review

Evaluate an AI-generated course draft and identify the most immediate governance risk before release.

03

Resume Screener Review

Assess AI-assisted hiring outcomes and recommend controls for fair, accountable screening.

+

Verified response

Revise weak incident responses, then justify a proportionate action through human impact, evidence, and accountability.

Learning design

Branch, Adapt, Verify

Learners examine mission evidence, make decisions, request intelligence when needed, and monitor a secure channel. Diagnostic performance changes the support they receive. Safeguard choices change when harm is detected, and weak incident responses must be revised before learners can verify their reasoning and complete a mission.

Progress, choices, and completion are saved on the learner's device so the experience can continue after a refresh.

LMS delivery

Ready for an LMS-friendly wrapper

The current course is a standalone web experience, but its interaction model can be instrumented for LMS delivery. A SCORM version would report bookmarking, completion, and score.

An xAPI or cmi5 implementation would additionally capture mission choices, safeguard decisions, support routes, response revisions, and verified rationales in a Learning Record Store.

Course continuity

Built to be handed off

Because the course was developed as maintainable source rather than a locked final product, a client delivery can include a SCORM-wrapped LMS package and a companion AI maintenance skill compatible with tools such as Codex or Claude Code. The skill can automate common content changes, validation, rebuilding, packaging, and handoff documentation—supporting course continuity while keeping review, LMS configuration, and publishing under client control.

Explore the field lab

Three systems. Three missions.

Move from abstract responsible-AI principles to decisions that ask what evidence matters, which safeguards are proportionate, and how an organization should respond when harm appears.

Launch the course Back to portfolio