Generative AI Use Policy

This policy was developed for Bridgewater Law, the same fictional disability law firm from my AI Email Workflow case study, and is also structured as a generalized methodology template usable for any organization. I worked with AI to draft the initial structure, aligned to the NIST AI RMF's four core functions (Govern, Map, Measure, Manage). I then applied judgment drawn directly from my own experience reviewing AI-assisted output in a federal disability determination environment — including real failure patterns I've had to catch and correct in practice: AI-generated narratives carrying details from one case into an unrelated one, unsupported facts introduced into a record, and severity levels that don't match the underlying documentation. The policy names these specific failure patterns and gives reviewers concrete guidance for catching them, rather than relying on generic "review AI output carefully" language.