In February 2011, “Jeopardy!” legends Ken Jennings and Brad Rutter appeared on a special edition of the show. Their challenge was to face a daunting, potentially unbeatable third opponent, an opponent that was highly intimidating despite having never appeared on the show—or slept, eaten or done anything else humans do for that matter. Their opponent was Watson, IBM’s artificial intelligence system. And prove unbeatable Watson did, with the final result (post-Final Jeopardy!) being Jennings, $24,000; Rutter, $21,600; and Watson, $77,147. In Jennings’ final answer, he wrote (paraphrasing a famous joke from “The Simpsons”), “I, for one, welcome our new robot overlords.”
Watson’s game show dominance manifested itself over 12 years ago. But a mere 12 months ago, on November 30, 2022, OpenAI’s ChatGPT system—referred to as a “large language model” not simply capable of Watson’s factual recitation, but also of being a virtual assistant with, well, virtually anything, from writing short essays, to coding, to extrapolating key ideas from lengthy texts—was unveiled to the public. Per Reuters, ChatGPT hit 100 million active users within an astonishingly fast two months. (For comparison, according to Visual Capitalist, it took seven years for the internet to reach 100 million users in the 1990s after its introduction.) Yet artificial intelligence is not a new concept, having first been introduced in 1951, when two scientists, Marvin Minsky and Dean Edmonds, developed what would be known as the first artificial neural network made up of 3,000 vacuum tubes that simulated our own brains’ neurons.
While it took a bit to get from 1951 to Watson (60 years), it took just over a decade to get to ChatGPT. Google scientist, inventor and futurist Ray Kurzweil coined the term for the increase in the speed of growth as the Law of Accelerating Returns, basically, that technology begets better technology, and that better technology can create exponentially better technology. In a 2001 writing about his Law, he posited that as a result, “we won’t experience 100 years of progress in the 21st century – it will be more like 20,000 years of progress (at today’s rate).”
So, what does any of this primer, though in part fascinating and in part Watson-level intimidating, have to do with transactional practice, or law at large? It’s simply that the tentacles of artificial intelligence will broaden and increasingly impact larger areas of the legal field, and faster than one might think. In the same breath that Watson—and other similar technologies over the last 10 years, like DeepMind’s AlphaGo and IBM’s Deep Blue (chess)—was what is referred to as “artificial specific intelligence,” that is, artificial intelligence that’s really good at answering trivia, or playing Go or chess, it can’t do anything else. It can’t reason, it can’t pick up a cup, it can’t negotiate purchase agreements.
Similarly, artificial specific intelligence (as the Business Law Update has noted in more detail in prior writings) has proven immensely beneficial in M&A transactions, especially those involving large amounts of diligence. AI programs like Kira, eBrevia and others are machine-learning software that identifies and analyzes contracts and documents, which saves attorneys and clients time and provides attorneys with accurate and distilled information. M&A attorneys can now have the software quickly review important contracts and datasets with a decreased fear of missing an important provision. The technology has proven essential for lawyers and clients by increasing efficiency and decreasing risk.
But AI’s development won’t stop at helping with diligence and document review—the next step is for AI to develop “artificial general intelligence,” which is AI that, broadly speaking, is far more capable of helping lawyers conduct research, draft documents and negotiate what is “market.” AGI is also more capable of learning trends, patterns and changes in the market. Further, it can help an M&A lawyer’s clients as well. The use of AI-driven data analytic software can help professionals and companies make informed transaction decisions before even speaking with an attorney. As further explained by the Institute for Mergers, Acquisitions & Alliances, AI has the ability to identify potential acquisition targets, assess each target’s alignment with strategic goals and evaluate potential risks during the transaction process, saving countless hours that were traditionally spent by clients on extensive research and analysis. Indeed, the IMAA stated that “AI algorithms provide insights into which targets are likely to align most effectively with the company’s strategic goals…[minimizing] the risk of pursuing acquisitions that may not yield desire outcomes.”
Kurzweil’s optimism may well be merited in terms of the increasing speed with which AI will continue to improve and broaden its ability to assist lawyers and their clients in M&A transactions. We will continue to evaluate the effect AI might have on the legal profession—staffing of deals, cost efficiency and volume of transactions that are feasible—as well as potential ethical risks (as happened this past June, when lawyers unwisely relied on ChatGPT’s responses that cited cases that did not exist). For now, we’ll encourage M&A attorneys to monitor these changes on a regular basis, and as AI developments emerge, actively consider how this technology can be beneficial to them and their clients and further enhance their practice. AI is here to stay and not something for attorneys to steer clear of; instead, attorneys should embrace its many uses and learn to communicate those benefits to clients. Not only is the use of artificial intelligence cost saving to clients, but its ability to analyze risk, enhance data processing, and streamline the transaction process will prove beneficial to the attorneys and law firms that embrace and leverage the technology. Mr. Jennings’s joking answer 12 years ago may well prove a lesson for practitioners and clients alike.
This newsletter may be reproduced, in whole or in part, with the prior permission of Thompson Hine LLP and acknowledgement of its source and copyright. This publication is intended to inform clients about legal matters of current interest. It is not intended as legal advice. Readers should not act upon the information contained in it without professional counsel. This document may be considered attorney advertising in some jurisdictions.
© 2023 THOMPSON HINE LLP. ALL RIGHTS RESERVED.