Three Concrete Ways Small Law Firms Are Using AI Right Now

Most AI coverage in the legal space runs toward the theoretical. What is possible, what might be disrupted, what attorneys should be prepared for. This is about what is actually happening in small firm practices today, in practices with one to five attorneys who do not have dedicated technology teams.

None of these uses are experimental. They are operational. They save time in ways that are measurable.

Status Call Preparation

A status call with a client is often a 30-minute conversation that requires 45 minutes of case review beforehand. The attorney needs to know where the matter stands, what has happened since the last call, what is coming up, and what questions are likely to arise.

Several small firms are using AI to generate that briefing document. They input the case notes, the recent correspondence, and the upcoming deadlines, and ask the tool to summarize the current status, flag open items, and identify anything the client is likely to ask about.

The output is not sent to the client. It is used internally by the attorney to prepare. The review step is fast because the attorney knows the file and can spot gaps quickly. The time saved is the 30 minutes of sifting through the file to build the picture manually.

Research Starting Points

AI tools are useful for identifying which cases and statutes are relevant to a specific question, particularly in practice areas where the attorney has strong foundational knowledge and is doing initial scoping rather than deep analysis.

The pattern that works: ask a specific, bounded legal question, review the output for the cases and legal principles it identifies, then verify each one independently using Westlaw, Lexis, or a jurisdiction-specific database. AI gets the attorney oriented. The authoritative research happens in the authoritative tools.

What does not work: treating AI-generated case citations as verified. The tools still hallucinate citations. The verification step is not optional.

First-Draft Correspondence for Recurring Matter Types

Practices that handle a high volume of a specific matter type produce a lot of similar correspondence. Status letters, document request letters, explanatory letters to clients about procedural steps. These follow predictable structures.

AI generates a serviceable first draft in a fraction of the time it takes to write from scratch. The attorney or paralegal reviews and customizes. The savings compound across a practice that handles 30 or 50 similar matters simultaneously.

The key to making this work is specificity in the prompt. Vague instructions produce generic output. A prompt that includes the client situation, the specific document being requested, the deadline, and the appropriate tone produces something worth editing.

What These Three Have in Common

None of them replace the attorney's role in the matter. All of them remove work that does not require attorney-level judgment. That is the useful boundary: AI handles the scaffolding, the practitioner handles the substance.

The firms using these tools effectively are not thinking about AI as a technology initiative. They are thinking about where time is going and whether it is being spent on the right things.