Extract from John Tredennick and Dr. William Webber’s article “Five Ways to Use ChatGPT for Investigations and E-Discovery”
Since its release in late 2022, ChatGPT has captivated the world, drawing a million users in its first five days and totaling over 100 million today. Whether you are fearful or excited, there is little doubt we have entered a transformational era, much like the invention of the printing press, the steam engine, cell phones, and the Internet itself. The extent and speed of ChatGPT’s progress (and of competitor systems) remains the only open question.
Like everyone else, we have read lots of flowery articles about the potential of ChatGPT to redefine the professional world. In this article, we intend to go beyond generalities by demonstrating five specific ways legal professionals can use ChatGPT (and its underlying GPT engine) to make ediscovery more efficient and cost-effective.
Using GPT and a basic search engine, we can show how legal professionals will soon take advantage of Large Language Models like GPT to streamline their ediscovery processes and better understand the key documents in their cases.
Prepare to be amazed.
Our Topic for This Exercise
Most ediscovery efforts start with a Rule 34 Request for Production. For this exercise, we will use one of the legal track topics from NIST’s annual Text Retrieval Conference (TREC), which was created in part to test the ability of different machine learning algorithms to find relevant documents