
Record an AI citation audit walkthrough
You'll have a 3-5 minute screen recording showing your real legal work (contract, memo, brief) and how you verify each AI-generated citation back to source material, with timestamps where you caught a
What you'll make
AI Citation Audit Walkthrough
- Open: the document and AI assistance used
- Category 1 check: fabricated case verification
- Category 2 check: jurisdiction and domain validation
- Category 3 check: cross-document contradiction hunting
- Close: errors found and time investment
Upload your 3-5 minute screen recording showing the verification process on your actual work product
The 3 parts
Each part ends with a check you can see.
Test Three Verification Approaches
Take one AI-generated legal document and verify the same citations using three different methods to discover what each approach catches versus misses.
Done when
You have tested the same 5-8 citations using basic validation, jurisdictional checking, and cross-document review, with documented findings about which errors each method revealed.
Build Cross-Document Contradiction Detection
Create and test a systematic method for finding when cited authorities contradict each other or undermine your legal argument.
Done when
You have a working process for checking cross-document contradictions that caught at least one contradiction AI missed, tested on real legal content.
Record Your Live Audit
Screen-record yourself running your tested verification protocol on real AI-generated content, capturing each error category you find.
Done when
You have a 3-5 minute recording showing verification of real legal content with timestamps for each type of error discovered and documented time investment.
Capstone: you put together your AI Citation Audit Walkthrough.
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
- Record a 3-5 minute walkthrough showing real citation verification on your own legal work
- Identify and demonstrate all three categories of legal AI hallucination in your verification process
- Have a tested workflow for catching cross-document contradictions that basic validators miss
- Document specific timestamps where you caught each type of AI error in your recording
Who it's for
- Lawyers using AI tools for research, drafting, or case analysis who need to verify outputs before filing
- Legal professionals responsible for work product quality when AI is involved
- Anyone producing legal documents where citation accuracy affects professional liability
What it won't cover
- Setting up screen recording software
- Legal research methodology without AI
- Malpractice insurance policy details
- Specific legal AI tool recommendations or comparisons
Skills you'll practise
- Multi-category hallucination detection
- Cross-document contradiction auditing
- Citation verification workflow design
- Legal AI output classification