
Compose a privilege-safe AI prompt library
You'll have a documented set of 5-8 AI prompts your team can reuse for contract review, discovery, and memo drafting—all tested with private or licensed AI tools that don't share data with third parti
What you'll make
Legal AI Prompt Library
- Privilege-Safe Prompts: 5-8 tested prompts with use cases
- Validation Checklists: how to catch each hallucination category
- Tool Requirements: which tasks need private AI vs specialized platforms
- Usage Guidelines: when to trust output vs when to verify
Structure as a practical reference guide your team can use immediately
The 3 parts
Each part ends with a check you can see.
Audit Your AI Tool Stack
Check which AI tools you're currently using and map their privacy policies against privilege requirements
Done when
You have a categorized list of 3-5 AI tools marked as privilege-safe, risky, or prohibited for legal work
Test the Three Failure Modes
Use AI on real legal tasks and categorize its failures as fabricated cases, wrong jurisdiction, or buried contradictions
Done when
You have documented examples of all three hallucination categories from your own AI testing, with specific failure cases identified
Build Privilege-Safe Prompt Templates
Create 5-8 specific prompt templates for common legal tasks that work with private AI tools and include built-in validation steps
Done when
You have 5-8 prompt templates tested on real work, each with documented validation steps and privilege safety measures
Capstone: you put together your Legal AI Prompt Library.
Where this helps
An elective: a single lesson that sits outside rtwo's job certificates and doesn't count toward one.
About this lesson
What you'll be able to do
- Run 5-8 tested AI prompts on real legal tasks with documented privilege safety
- Classify AI output failures by hallucination category to catch malpractice risks before they ship
- Have a reusable validation checklist for any new legal AI workflow
- Know which prompt patterns work with private AI tools vs which require specialized legal platforms
Who it's for
- Solo practitioners and small firm lawyers who want to use AI but can't risk privilege breaches
- Legal teams at companies who need AI efficiency without exposing confidential information
- Anyone doing contract review, discovery, or memo drafting who's been burned by AI hallucinations
What it won't cover
- Setting up private AI instances or enterprise contracts
- Replacing legal research platforms with general AI tools
- Automating final work product without human review
- Constitutional law or specialized practice areas beyond contracts and business law
Skills you'll practise
- Hallucination pattern recognition
- Privilege-safe prompt design
- AI output validation
- Legal workflow decomposition