UPDATE: We’re grateful for the positive response to this article. It has been shared on JDSupra and EDRM. Thank you for the support!

Merlin’s CEO, John Tredennick, and Chief Data Scientist, Dr. William Webber, wrote a follow-up article to “What Will Ediscovery Lawyers Do After ChatGPT?” on the use of ChatGPT in ediscovery. It has been published in the February 22, 2023 issue of  LegalTech News and Law.com, a publication of American Lawyer Media. 

The article shared results from the authors’ research using the new Large Language Model (LLM) algorithm to address common ediscovery tasks such as crafting a Boolean search, document summarization and even first-pass review. It was the first article of its type since the public release of CHATGPT by OpenAI just six weeks earlier. 

ChatGPT is an AI tool capable of answering complex questions and generating a conversational response. (Indeed, it reportedly passed a Wharton MBA exam and the three-part U.S. Medical Licensing Exam.) It generates text by iteratively predicting the most likely word or sequence of words to appear next, given the words already written. Imagine continuously hitting “TAB” on a much smarter autocomplete.

The underlying software is powered by a deep-learning AI model called GPT-3 (soon to be superseded by GPT-4). It is a powerful language processing AI model trained by “reading” around 350 billion words crawled from the Web, Wikipedia, and book collections. The deep learning model extracts about 175 billion parameters from this data, requiring a huge amount of computational power to do it. 

“We wrote this article because the world is talking about this new technology and some are fearful of its reach,” Tredennick noted. “With the recent release of ChatGPT, even legal professionals are starting to wonder about job security.  Can ChatGPT replace first-pass reviewers? Can it answer legal questions? Will it one day represent a client in court?” 

We asked ChatGPT directly and got this answer, “ChatGPT can be trained to identify relevant documents and classify them based on their content, reducing the need for human review.”

Ultimately, the authors tried to answer the question, will ChatGPT replace ediscovery review teams? Read the full article on law.com  or download a PDF from our site to see their answer.