Artificial Intelligence is Seeping into the Legal Field

By: Emily Ashley

Yavar Bathaee, a partner at a law firm that heavily leverages Artificial Intelligence (AI) capabilities, wants to replicate the judgment of a seasoned trial lawyer who has been practicing for 30 years. His firm, Pierce Bainbridge, is experimenting with AI in a variety of ways, such as determining whether a Federal Rules of Civil Procedure 12(b)(6) motion to dismiss will survive a challenge. 

AI is subtly seeping into every aspect of our lives and the legal field is no exception. 

When we talk about legal AI, “often, it’s not even AI — it’s automation of routine tasks that otherwise would be carried out by a person with lots of experience,” said Colleen Chien, Professor of tech law at Santa Clara University School of Law and advisor to ClearAccessIP, an AI-driven patent analytics and enterprise software firm. 

Lawyers are already leveraging automation to perform tedious work, such as document review or completing paperwork, said Chien. But AI goes deeper than automation, it involves machines that are actually trying to mimic human intelligence and behavior. The seemingly limitless capabilities make it hard to predict how these tools will fit into the legal landscape, said Chien. 

“When the [tool] is learning from data and it’s fully-trained, it can arrive at decision boundaries that no one ever intended,'' said Bathaee.

A number of AI tools are already in use. ROSS Intelligence, a legal research platform used by firms across the country, is using AI to expedite legal research. ROSS provides a platform that uses natural language processing capabilities to reduce the time and effort it takes to refine search queries. They also offer a document analyzer where users can upload memos or briefs to determine if their citations are still good law. ROSS is working towards running searches that will reflect how individual judges have historically ruled on various issues. Bathaee envisions a future where lawyers and judges are using AI to keep judgments consistent with precedent. 

“The question then becomes predicting the judgment of the person making the decision more directly from what they’ve done and what drives them, than playing this jurisprudence game where I read a case and I distinguish the case’s facts — that’s going to become less important than saying, ‘Ok, Judge X usually has the following outcome on the following issue,’” Bathaee said. “You’re looking at what judges actually do and not what they’re supposed to do.”

Additionally, AI can help lawyers work more efficiently under attorney-client privilege, said Bathaee. 

“AI is a great way to preserve privilege,” Bathaee said. “AI can look through privileged documents, make privileged calls, or even look through opposing counsel’s privileged documents and make determinations without a human being breaking the privilege.”

Workforce Evolution 

Many wonder how AI will impact the legal job market. Joanna Goodman, AI researcher and author of Robots in Law: How Artificial Intelligence is Transforming Legal Services, is not concerned. “In Silicon Valley, you’re client-facing. [AI] will actually allow junior lawyers to see more how the client works rather than sifting through documents.” 

Bathaee said that “the tools have gotten so powerful that they can leverage themselves, so they essentially have the benefit of having an entire staff of junior lawyers. There’s two sides to this coin. On one hand, you probably won’t find that easy job [at a firm] doing grunt work, but on the other hand, you can compete directly with them without having a giant leveraged staff. The idea is that if you’re younger and you’re less experienced, you can actually acquire that experience without the barriers of entry we normally encountered 10-20 years ago.”

Goodman warned that when it comes to AI, “law is not special, law is late to the game.” She encouraged innovation, pointing out that “process is a barrier to progress … I don’t think the legal tech sector realizes that by saying, ‘Oh, we don’t want to do this, we want to go back to boring and look at the data,’ they don’t realize that someone else will get in first because tech is fast-moving.”

To complicate this risk, the Task Force on Access through Innovation of Legal Services, a committee appointed by CA State Bar Board of Trustees, met this past summer and recommended that the state allow non-lawyers and technology-driven legal systems to deliver legal advice and services. 

If the door opens for non-lawyers and AI to act as lawyers, the Big Four accounting firms — PricewaterhouseCoopers, Ernst & Young, KPMG, and Deloitte — are primed to enter the fray as they have started investing in legal tech and purchasing products, such as AI contract review platforms, to begin providing those services to their clients. Google is also entering the fold, offering a product called Document AI that analyzes legal documents. 

Embracing the Change

With an evolution on the horizon, adaptability is a natural concern. “Those who are open to its possibilities and how it can actually improve and make the legal system work better for everybody are going to be in much better shape,” Chien said. “Lawyers also need to be aware that people are unhappy with the legal system. They find that it’s broken. The more resistance there is in the legal industry to change, the faster the disrupters who are unhappy with it will try to change the system fundamentally. I think it’s in everybody’s interest to embrace change and try to make the system work better for everybody, not just for lawyers and making more money. Those who are concerned about the future of the industry should try to think about how we can add value.”

Lawyers do not necessarily need to master programming to embrace AI, but familiarity will help. 

Athena Fan, legal technologist and founder of the start-up Bite Size Legal, agreed. “You need to know enough about engineering or programming to at least talk to the people who are programming the software. Once you understand the process it takes to code, it empowers you to talk to your programmers,” she said. Bite Size Legal is a legal chatbot meant to resolve legal rental issues and is currently in early Beta Testing. 

“Being able to write very basic computer programs, build models, being literate in data science will give you a huge edge,” Bathaee said. He is optimistic: “In a few years, people will have the baseline ability to write a Python script that reads some data and builds a neural network that’s trained on it and makes decisions,” he said. 

Chien agreed that coding is becoming a norm and points out that “more and more of my students have that background coming to law school. I feel pretty confident that for the next generation, that’s not going to be a real differentiator.”

(Editor's Note: This article was originally published in the October 2019 [Volume 50, Issue 1] version of The Advocate.)