AI System for Preparing Case Law Research

The problem

In day-to-day operations, a narrow query may miss relevant decisions. When the research note, chronology, concepts and case law database citations are spread across different screens, documents or team records, the issue may not be noticed in time.

This fragmented data delays preparation of the sourced list of candidate authorities and forces staff to search for supporting material again by hand. Before deciding, the research lawyer must complete missing fields and verify each inconsistency individually.

A common exception is that a decision may have lost relevance because of later legislation or case law. If this context is not checked, an incomplete or incorrect automation could cause an irrelevant decision to be treated as supporting authority, so the output cannot be acted on directly.

What we set out to improve

• Share of sources the lawyer considers worth researching
• Share of outputs corrected or rejected by the research lawyer
• Time between the source record and the review output

How the system works

1. 01 Define the review scope and the authorised research lawyer role
2. 02 Collect the inputs: research note, chronology, concepts and case law database citations
3. 03 Check the source identifier, date, version and data freshness
4. 04 Extract concept, date and jurisdiction filters
5. 05 List citations with their sources and currency
6. 06 Check the exception: a decision may have lost relevance because of later legislation or case law
7. 07 Place low-confidence or conflicting records in a separate human review queue
8. 08 Write the source identifier, rule, output and timestamp to the audit trail
9. 09 Have the research lawyer approve, correct, defer or reject the output

Methods we used

• Query expansion and source retrieval
• Source, format, version and required-field checks for the research note, chronology, concepts and case law database citations
• A missing-data alert instead of an estimate when inputs are absent or conflicting
• Separate handling of exceptions and the main rule
• Joint retention of the source identifier, rule result, sourced list of candidate authorities and human decision

Where people stay involved

Data and security

Access to the research note, chronology, concepts and case law database citations is limited to the minimum permissions needed for the task. Identifiers and sensitive fields are masked where possible, and only the necessary section is sent to the model rather than the full raw file. Access, output and the research lawyer's decision are logged; production use does not begin until the organisation approves the retention period and data location.

Who this suits

Frequently asked questions