Case Decode AI

An AI tool for law students that reconstructs legal judgments into case digests to improve the efficiency of understanding chains of judicial reasoning. It addresses key limitations of existing AI tools: over-compression, loss of nuance, and hallucination risk.

It does not simply reduce reading into simplified summaries at the expense of developing essential skills such as legal comprehension and reasoning. Instead, it supports learning by preserving accuracy, restructuring information into a format that is easier to understand, and providing convenient ways to verify information against the original source material

1. Reorganisation rather than summarisation - To reduce unnecessary compression and preserve important nuances
2. Extraction of chains of legal reasoning - To show how judges moved from facts and legal principles to conclusions
3. Sole source derivation - To ensure outputs remain grounded in the uploaded materials and minimise hallucinations
4. Explanation and follow-up question feature - To provide quick access to definitions and plain-language explanations through highlighted text
5. Find in Source feature - To allow users to verify generated content efficiently
6. Verdict pills (YES / NO / UNRESOLVED) with brief summaries - To provide a quick understanding of the overall direction of the case
7. Case facts section - To give users the factual background needed to understand the reasoning
8. Issues section - To clearly outline the legal questions being considered
9. Reading time estimate - To give users a sense of the scope and manageability of the reasoning map before reading