# Case Analysis with AI: The IRAC Method

Law students learn IRAC in their first week: Issue, Rule, Application, Conclusion. It is the backbone of legal analysis. What most people do not realize is that IRAC also happens to be a perfect framework for prompting AI to summarize cases.

The reason is simple: AI without structure produces narrative summaries that read like Wikipedia entries. They tell you what happened but not why it matters. IRAC forces AI to extract the elements that are actually useful for legal work — the precise issue decided, the rule applied, how the court applied it, and the holding.

This matters because bad case summaries are not just unhelpful — they are dangerous. If AI misidentifies the holding, or conflates dictum with a binding rule, you could cite it in a brief and face sanctions. This lesson teaches you to get summaries that are reliable.

The IRAC Prompt Framework

Bad Prompt ` Summarize this case: [paste 30 pages of opinion]

Output: A chronological narrative that buries the holding in paragraph four and mixes facts with analysis. You have to re-read the entire summary to find what you need.