AI in the legal industry is here to stay. If you have read any articles lately about legal technology, chances are you have come across a few titles devoted to artificial intelligence; it’s hard to miss the pervasive impact AI is having across virtually every industry.
Artificial intelligence is the topic du jour in legal technology these days. Whether you are an attorney, a paralegal, or you work in some other legal support role, the seemingly rapid rise of new tech may have you feeling hopeful or fearful (or somewhere in between) about your future and the role of AI in the legal profession. Or perhaps all the talk about AI simply leaves you bewildered.
However you may feel about artificial intelligence in the legal industry, there’s no doubt it is already prompting a rapid and fundamental transformation.
Legal professionals who want to remain relevant must understand this new technology and stay current on its development to succeed. Those who do not will likely lag behind and find it increasingly difficult to maintain their career and business goals.
In this article, I’m going to discuss some basic concepts about artificial intelligence and the changes AI technology will likely bring to the legal profession.
What is artificial intelligence?
The concept of “artificial intelligence,” or “AI,” has been around since the late 1950s, but many people still associate it with science fiction. The phrase for many of us will conjure up images of Hal 9000, the rogue computer from 2001: A Space Odyssey.
Of course, there’s nothing sci-fi about AI’s applications today. Artificial intelligence broadly refers to computer systems that can perform tasks typically requiring human intelligence.
AI aims to create machines that can think, learn, and make decisions similar to humans by using computer programs powered by algorithms.
An algorithm is a set of step-by-step instructions a computer can follow to solve problems or complete certain operations; it is like a recipe guiding a computer through specific steps to achieve a desired outcome.
Using algorithms, an AI-enabled computer is capable of performing certain tasks that would otherwise require human thinking, problem-solving, and decision-making skills.
Machine learning
Another term that is important to understand when discussing AI is “machine learning.”
Machine learning refers to computers powered by algorithms capable of processing large amounts of data quickly. These computers can identify patterns and anomalies in data, which makes it appear as if they “learn” over time.
Machine learning is already deployed in almost all sectors in the U.S. and elsewhere, including manufacturing, finance, and law, for example, to improve efficiency, among other tasks.
Deep learning
A particular sub-category of machine learning is called “deep learning,” which relies on an interconnected web of neural networks. These networks allow computers to not only spot patterns, but to engage in human-like reasoning based on the incoming data.
This type of AI is already being used in fraud detection, and it is vital to autonomous vehicles, which are required to rapidly process huge amounts of data in a dynamic environment and make decisions based on what the computer inside the vehicle perceives.
Regarding deep learning AI in the legal industry, it has already been found to have credible applications in contract analysis, legal research, and e-discovery, among other things.
The ability of deep learning models to analyze massive data sets (which law firms more often than not have) makes them ideal for assisting legal professionals.
Large language models and ChatGPT
By now, most people have heard of ChatGPT, even if they have not interacted with the technology. ChatGPT is an automated tool that makes use of a particular type of deep learning technology called large language models or “LLMs” (not to be confused with the advanced law degree).
LLMs focus on recognizing, understanding, and processing human languages, whether spoken or textual, and then reproducing text in a human-like fashion.
One source describes how ChatGPT was programmed by inputting massive amounts of text data from the Internet, including articles, books, websites, and other sources.
It was then trained to recognize relationships between words, phrases, and sentences, allowing it to understand language in context and generate coherent responses to user-generated prompts.
A version of ChatGPT is currently free to the public, and I would encourage you to visit the website and try it out to get a sense of what it does.
As an industry, legal services depend entirely on the use of language. After all, as the novelist Franz Kafka is attributed as saying, “A lawyer is a person who writes a 10,000-word document and calls it a brief.”
Because of its focus on comprehending and replicating human languages, LLM technology through tools like ChatGPT represents the type of artificial intelligence most capable of transforming virtually every aspect of daily law practice operations and management.
Let’s use Kafka’s quote as a reference for how AI could revolutionize just one task lawyers perform. A lawyer might spend several days or more writing a 10,000-word legal brief. The brief’s quality would depend entirely on the lawyer’s education, training, and skill.
I asked ChatGPT to provide an average timeframe it would it take to write a a 10,000-word legal brief. While the answer I received was predicated by disclaiming language advising me to consult with a human lawyer, the answer it ultimately gave was 30-45 minutes.
Would ChatGPT’s work be a quality, reliable brief one could submit to a court? Absolutely not. Many readers have undoubtedly heard about the New York lawyers who were sanctioned by a federal judge for submitting a legal brief produced by ChatGPT that contained citations to non-existent cases.
Even in that case, however, the judge remarked that he might not even have sanctioned the attorneys solely for using ChatGP to write a brief; he sanctioned them because they lied to him about it when they got caught.
As it currently exists, AI does not have the skills even a mediocre attorney possesses in writing briefs or motions… yet.
We must not forget, however, that the technology is still in its infancy. Moreover, brief writing requires considerable knowledge, creativity, and analytical reasoning. Many tasks performed in law practices every day do not require the same degree of high-level thinking.
Even in its infant iteration, AI is already capable of performing at least some of these tasks faster and with greater accuracy than humans.
In other words, while those New York lawyers may have tripped over themselves using ChatGPT, there’s certainly a case to be made that in the coming years, its growing sophistication will allow it to be used as an invaluable legal tool.
What does AI mean for the legal profession?
At this point, many readers may be asking themselves this legitimate question. As a 2018 Pew Research Center study revealed, 65% of Americans were fearful and uncertain about the effects automation will have on the workforce.
These fears have only increased in 2023, with many employees concerned about being replaced by AI, even as they increasingly use these tools.
Despite these concerns, no one seems to know a definite answer about AI’s effects on the legal profession in particular. If researchers and industry experts know the answer, they have so far been reticent to provide it.
Instead, most are more comfortable extolling the positive attributes of AI in law, such as increased efficiency and productivity, better work-life balance, and greater access to justice.
As laudable as these goals are for our profession, these also tend to be buzzwords and catchphrases that can mask the truth of what lies in store.
The time to be honest and prepare oneself is now. If the past is prologue, AI will likely result in job losses across all economic sectors, including the legal field. New technologies have for the most part, at least in modern times, resulted in the elimination of millions of jobs.
A 2019 Oxford Economics study entitled How Robots Change the World, predicted that 20 million manufacturing jobs will be lost to robots by 2030. The study also predicted robots would steadily gain a greater foothold in various service industries, particularly as AI advances, and this will likely result in job losses for less skilled workers and may contribute to increased income inequality.
A glimmer of hope for legal professionals can be found in this quote from the study’s authors, which I would like to allude to:
“It will be difficult for machines to replace humans in service sector occupations that demand compassion, creativity, and social intelligence. Physical therapists, dog trainers, and social workers are likely to remain secure in their jobs… even if truckers and warehouse workers see the future of their jobs jeopardized.”
In other words, the legal profession distinctly requires workers who are creatively minded and detailed thinkers, as problem-solving is a key aspect of the job of paralegals and attorneys. In this regard, AI in the legal industry is more likely to serve as a legal aid than it is an outright replacement for many professionals.
Not surprisingly, Gartner, Inc., an American technology research and consulting firm, paints a rosier picture, predicting that AI will create 2.8 million jobs while eliminating only 1.8 million. Statistics such as these are misleading, however, because these do not mean anyone’s current occupation is immune from elimination by AI.
Who’s most at risk from AI in the legal industry?
The legal professionals who are most at risk from AI in the legal industry are those who perform relatively mundane, repetitive tasks. If you are part of the legal support staff at your firm, and your job consists primarily of routine or repetitive tasks, your job is at risk.
If you are an attorney, and your day’s work is typically comprised of legal research, contract review, the preparation of transactional documents, or certain in-house tasks related to e-discovery, you are at risk.
People working on tasks like these will probably see their work drastically curtailed, if not eliminated completely, due to AI automation. In addition, generalist attorneys are more at risk of seeing at least some of their work lost to automation than specialists.
In 2018, before ChatGPT technology was released, a competition was held between AI technology and twenty seasoned transactional attorneys. The competitors were tasked with reviewing 153 paragraphs of contractual language to spot errors and allocation of risk.
The lawyers completed the task with an 85% accuracy rate, with the longest performance time for any lawyer at 156 minutes. At that rate, even the slowest attorney was reviewing almost a paragraph per minute.
As impressive as that result may seem, AI completed the task with a 94% accuracy rate, and it did so in 26 seconds. With the latest iteration of ChatGPT (GPT-4) able to pass the bar exam with a score nearing the 90th percentile, there may be few limits to the disruption this technology will bring to the practice of law.
If you are a legal professional and are concerned about the security of your position because of AI, then the time to review and renew your skillset is now.
AI in the legal industry in 2024
According to recent studies, lawyers appear to be optimistic about the potential that Gen AI offers for the future of the legal profession, but there is also some natural skepticism.
Reuters’ 2024 report on the state of the legal industry lays out three potential scenarios that may unfold as Gen AI evolves more fully, each with varying degrees of impact. These scenarios address how law firms serve their clients and the ultimate benefits that Gen AI could bring to the legal industry.
The report doesn’t provide detailed information on each scenario, but it emphasizes the need for law firms to be prepared for changes and potential adaptations in response to the evolving role of artificial intelligence in the legal field. Overall, it recognizes Gen AI as a significant factor that will shape the future landscape of the legal profession.
There are also significant shifts that pose challenges to traditional law firm business models, including the impact of generative AI.
The legal market showed improvement in 2023, but the report emphasizes a “sorting out” process where firms are testing various strategies.
Notable changes include legal work moving towards smaller firms, increased budget caps by clients, and a shift from transactional to counter-cyclical practices.
Rising rates are offset by lower realization and higher expenses, challenging profitability. The report emphasizes that successful law firms are those adapting faster to market realities.
Looking ahead to 2024, corporate work is expected to rebound, but direct expenses remain a burden. Generative AI is identified as a potential game-changer, with scenarios ranging from enhancing client value and firm profits to clients deriving disproportionate benefits.
The report urges law firm leaders to recognize the evolving landscape and emphasizes the need for openness to new ways of delivering legal services in the changing market.
Are law firms prepared for cyber threats?
The confidence in the readiness of both in-house counsels and law firms to handle digital threats from online adversaries is growing, as revealed by the findings of Bloomberg Law’s Legal Ops & Tech 2023 survey.
Almost half of all lawyers (46%) strongly affirm that their organizations are prepared to respond to confirmed cybersecurity threats like ransomware or data breaches. This marks a notable increase of 10 percentage points compared to the previous year’s survey, indicating a growing number of lawyers gaining confidence in their organizations’ cyber defense capabilities.
The disparity between 2022 and 2023 is particularly evident among in-house counsel, with 68% strongly agreeing in 2023 compared to 46% the previous year — a substantial 22% rise. Among law firm lawyers, the “strongly agree” responses increased by 13 percentage points, from 30% to 43%.
Conclusion
The rapid rise of artificial intelligence, particularly in the form of deep learning and large language models like ChatGPT, is transforming the legal industry.
While AI holds promise for enhancing efficiency and access to justice, it also raises concerns about potential job displacements. Legal professionals must adapt to this evolving landscape by embracing AI as a powerful tool to augment their expertise rather than replace it.
Creativity, problem-solving abilities, and social intelligence remain invaluable assets that AI cannot fully replicate.
As AI technology in the legal industry continues to advance, finding the right balance between human expertise and AI assistance will be critical to weather the gathering storm in the legal profession.
