AI is not a lawyer, but it can be a powerful force multiplier for litigators

Litigation will always require a lawyer’s experience and judgment. To a develop trial strategy, attorneys spend a lot of time reviewing transcripts, building outlines, creating arguments, thinking of counterarguments, and creating exhibits. A secure, enterprise version of ChatGPT, which prohibits OpenAI from using or retaining client data to train ChatGPT models, can quickly and effectively streamline this work for litigators. Fennemore is on the leading edge of using this secure, closed model.

ChatGPT’s responses are helpful, but not perfect. That is where a lawyer’s judgment comes in to evaluate ChatGPT’s work and decide the best step forward for the client’s success.

Here are some practical applications of an enterprise model of ChatGPT that is secure, adds depth and saves time; a win/win/win in litigation.

1. Mine trial and deposition transcripts for pinpoint cites and argument support. Upload deposition or trial transcripts to a secure workspace and ask ChatGPT for targeted pulls on a specific issue with page:line citations. Here’s an example prompt: “Using the uploaded transcripts only, find every place where Witness A (1) admits receiving the April 12 email, (2) discusses the pricing change, or (3) uses the phrase ‘standard practice.’ Return a table with topic, date of testimony, page:line, exact quote, and a one-sentence note on how it supports our argument.”
2. Build deposition and cross prep: funnels, exhibits, and impeachment hooks. If you provide the sources, ChatGPT can build question funnels (broad → narrow), propose exhibit sequences, and draft impeachment questions tied to prior testimony. For example, you can ask ChatGPT: “Goal: establish that the CFO knew the revenue-recognition problem before the financing round. Using Exhibit 12 (board deck) and the CFO’s prior depo excerpt (pages –), draft (1) the admissions to lock, (2) a question funnel, (3) the exhibit order, and (4) impeachment questions using only quoted prior testimony. Include citations to the exhibit page and deposition page:line.” Some, and in many situations, most of what ChatGPT responds with is unusable, in my opinion. I usually come away with several new ideas that I had not thought of before.
3. Stress-test a motion or brief like a skeptical judge. ChatGPT won’t replace legal research entirely, but it can help you think of counterarguments that you may be too entrenched on one side to see. The following model prompt with uploaded briefing would provide enough detail to deliver strong counterarguments to build from: “Here is our draft motion and statement of facts. Act as a skeptical trial judge. Identify the three weakest arguments and why, the facts needing stronger record support, and any overstatements that hurt credibility.”
4. Create punchy introductions and conclusions in a motion. I don’t let ChatGPT dictate the arguments to be made in a motion or response. However, once I have finished the arguments, I can create a contextual prompt, upload my motion, and utilize ChatGPT to draft a forceful introduction and conclusion with a call to action. Here is a prompt I recently utilized: “Using the arguments included in the attached motion, write a powerful and insightful conclusion that will persuade the reader to dismiss the plaintiff’s complaint with prejudice.” I will even ask ChatGPT to write an introduction or conclusion several times, perhaps with a slight change of tone (e.g., more straightforward or easier to read), and then I may select portions of each response to craft my own introduction or conclusion.
5. Financial or Bank Statements Review and Analysis. What would take me hours takes minutes when I upload financial or bank statements into ChatGPT. As a commercial litigator, working with complex financial data is an everyday task. ChatGPT makes synthesizing data quickly seamless. Recently, I received a complicated wire transaction report from Bank of America. The data had no headings! I asked ChatGPT to sort the data into an Excel file. Impressively, it suggested the appropriate headings, and I was able to see where money went to and from effortlessly. I verified the information and turned the synthesized data into a Rule 1006 summary to be used at trial. What would have been an unintelligible mess turned into a ready-to-use trial exhibit that I could verify and finesse in minutes.

ChatGPT is not a substitute for my or any litigator’s judgment, and it never should be. But used correctly, it is a powerful force multiplier. It frees lawyers like me, who deal with complex and voluminous data, from the grind of organizational work and gives us more time to do what clients pay for: think strategically, assess risk, exercise judgment, inject passion and finesse arguments.

Moving forward, litigators who will have the greatest advantage are not the ones who resist AI or blindly rely on it, but rather the ones who learn how to grow with it thoughtfully, skeptically, and ethically. In modern litigation, efficiency, heightened organization, healthy skepticism, and added depth are not luxuries, they are a competitive edge.

Emily Ward is a business litigation attorney focusing on complex and high-dollar business disputes in Arizona. She concentrates on business litigation, public policy, and appeals, litigating high-stakes cases for both plaintiffs and defendants in federal and state courts across the country. Reach her at eward@fennemorelaw.com.