Susskind’s AI Bet: Too Far or Not Far Enough?
The famed author and lecturer has made some bold bets about the future of law. His views on AI have sparked much conversation and debate in the legal community.
Four years ago, Richard Susskind published the first edition of “Tomorrow’s Lawyers: An Introduction to Your Future.” With the rapid changes in the legal profession, tomorrow is now today.
The second edition of “Tomorrow’s Lawyers” focuses more sharply on how artificial intelligence, alternative business structures, low-cost law firm service centers, legal tech startups and evolving in-house roles are changing the way legal services are delivered and how law schools are educating students to meet those changes.
To that end, ALM during October is publishing excerpts across several of our brands from the second edition to spark thought and conversation about the industry’s future among the legal profession’s leaders. ALM editors and reporters have solicited reactions — positive and negative — to Susskind’s ideas from law firm chairs, top legal educators, general counsel, law students and industry analysts to get their take.
Author Richard Susskind has made a name for himself through bold predictions about the future of the legal profession. And in much of his writing and lecturing, Susskind has focused on technology’s role in the delivery of legal services, noting in the second edition of Tomorrow’s Lawyers that by the 2020s, the legal profession will undergo “transformational” changes related to artificial intelligence (AI). By 2036, he posits, “It is neither hyperbolic nor fanciful to expect that the legal profession will have changed beyond recognition.”
But such predictions aren’t without challenge. Reviewing The Future of the Professions, a book Susskind co-authored with his son and fellow researcher, Daniel Susskind, University of Maryland professor of law Frank Pasquale writes that the Susskinds “cherry-pick” a study on computerization’s impact of jobs. And while Pasquale notes “experts differ on the real likelihood of pervasive legal automation” he notes that the study’s authors place the “risk to lawyers” of automation at less than four percent, though project that paralegals “are in much more danger.”
Others agree with Susskind’s predictions but take fault with their interpretation. As Ryan McClead, VP of client engagement and strategy at Neota Logic, told Legaltech News, “‘We are in a state where the public perception of artificial intelligence is magic, and the reality is it’s not going to be for a long time. It will not be magic by 2036. AI is work it requires work up front no matter what type of tool you’re using.”
As for that specific date of 2036, McClead noted, “That’s a safe bet because if he’s wrong, nobody’s going to fault him for it. Most industries in 20 years are going to be significantly different than they are now. I think he’s right, but he made a safe prediction.”
LTN recently spoke with experts at the intersection of AI and the law to get their takes on Susskind’s predictions about AI in Tomorrow’s Lawyers.
A Gradual Yet Underestimated Change
The media have made much of AI in recent years, with grandiose claims of replacing lawyers with robots. The result is numerous misconceptions around the technology’s potential. Susskind agrees that many claims “overstate the likely [near term] impact of AI,” though he is a bit more nuanced in his view, stating that while “extravagant,” many claims understate the long term impact.
Legalmosaic CEO and founder Mark Cohen takes a similar approach, noting that while technologies like the DoNotPay chatbot are useful, they aren’t about to replace the human lawyer.
“I think that a lot of it is AI is the shiny new toy, so everybody’s is focused on it. There’s lot of sensationalism. Most people don’t even know what AI is, much less what it can do,” said Cohen, who is also a fellow at Northwestern University’s Pritzker School of Law and has worked as a civil trial lawyer and managing partner at Finley Kumble. “I don’t think AI is going to replace lawyers wholesale. Will it perform certain high volume, relatively low value functions that lawyers presently do? I’m sure. It already it is. But ultimately are robots going to try cases? No. Will robots help with predicting outcomes? Likely so.”
In Susskind’s estimation, technological capabilities will inevitably lead to machines conducting increasingly more legal tasks “historically regarded as the unique preserve of legal practitioners.” In some ways, this is starting to happen: Law firms like DLA Piper are employing AI technology for due diligence in M&A work, while ROSS Intelligence, a legal research tool built on IBM Watson, is being adopted by firms like Womble Carlyle Sandridge & Rice and Baker & Hostetler.
Andrew Arruda, founder of Ross Intelligence, told LTN that claims around AI “aren’t necessarily overstated,” but that the media and some companies have gone overboard with boasting AI’s capabilities.
“When you talk about AI, you have to realize that its integration in software will become ubiquitous. What we will come to expect as status quo from our software will really blow away today’s tech, but we won’t call it AI, we will just call it software,” Arruda told LTN. “There will always be a need for the human component of legal work, and when that isn’t the case, society will have changed so fundamentally that the concept of what a lawyer is might not even make sense.”
Legal Services Overturned
Yet Susskind suggests that technology is changing the concept of legal work, along with the lawyer’s role in preparing that technology to execute it. He notes that AI in legal presently rests in the terrain of making predictions, answering questions and document identification — a far cry from a robot lawyer — but says it’s “vital that we ask whether people or systems will be doing legal work in decades to come.”
Cohen said that such an assertion actually begs the more important question of, “What is a lawyer, and how does a lawyer operate?”
“It used to be that lawyers were the only ones that decide what is legal and what’s not. Now increasingly, other people are making that determination,” Cohen said, referring to alternative legal services providers. “I do agree with Richard to the extent that there was this urban myth that everything lawyers did required lawyers and that virtually everything lawyers did was bespoke. I totally agree that was and now has been largely debunked as just a myth.”
Indeed, companies like Axiom, Atrium and LegalZoom are cutting into the market traditionally cornered by law firms and solo practitioners. At the core of this change are still people. However, that hasn’t stopped much legal work form being at least partially replaced by machines — think document review. Susskind writes that many lawyers believe that because “computers cannot think or feel,” they “cannot exercise judgement or be empathetic.” This, he continues, rests on the “AI Fallacy, the view that the only way to get machines to outperform the best human lawyers will be to copy the way that human lawyers work.”
McClead said that while “machines can absolutely outperform humans at specific tasks … when you start training a machine learning tool, it’s really dumb. When you finish training it, it’s really dumb, but it does that one thing really well and really fast and usually better and more accurately than a person does. It takes a lot of work to do that. You end up with machines and tools that augment what lawyers are doing, but they’re not replacing it.”
As to whether advances in big data and machine learning validate this statement, Cohen said that such a statement needs to be more specific.
“Law in the highest form is the persuasion business, and I don’t care whether it’s trial work or negotiating a contract, law is persuasion,” he said. “The person who is the most effective lawyer need not have the highest IQ, but he or she must have a combination of IQ and EQ, and by the way be smart enough to understand when they should probably turn to machines to get data. But they’ve got to be very careful in terms of what data they ask for.”
Lawyers In The Future
Susskind reasons that given computing power in other professions as well as advancements in law, machines are sure “to eat into lawyers’ jobs,” suggesting that those beyond “the best and the brightest human professionals” could be replaced by machines. Yet, he writes that as early as the 2020s, lawyers will likely be required to “undertake different work.” For their careers, he says that they “should plan either to compete with machines” — basically find legal jobs that prefer specifi humanc skills over AI — or “build the machines.”
Paul Carr, president at Axiom, said that he interpreted this to be lawyers training machines.
“‘I don’t think en masse the legal profession are going to become technologists. That seems hard to believe,” he explained. “But certainly whether it is codifying expertise in rule-based systems or whether it is training machine-based learning technologies, that cannot happen in the absence of subject matter expertise and technical understanding.”
Meanwhile, Cohen agrees with Susskind to the extent that law will become more interdisciplinary. However, he added, “Let’s face it, a lot of people go to law school because they didn’t do well in math or science. So I don’t think it’s reasonable to conclude that a wide number of young lawyers are going to be AI developers.”
Yet Susskind ultimately says that far into the future, “it is hard to avoid the conclusion that there will be much less need for conventional lawyers.”
This assertion, Carr said, is “deliberately and necessarily provocative.”
“So long as you’ve got hellishly complicated legal regulations and frameworks in every single corner of business, whether it’s employment or commercial or environmental, you need lawyers. And so long as you have disputes, arbitrations and disagreements between companies and people, and companies and companies, you’re going to need lawyers.”
McClead agrees, noting, “Like it or not, we will always have a conventional need for lawyers. Much of what lawyers do will always be done by humans. They will be enhanced and augmented by technology, but human lawyers will probably always be at the heart of any technological solutions, advocating for their clients.”
Arruda, like McClead, says that much of the future of legal relies on lawyers and technology working in tandem rather than one over the other.
“I think that if tech is used in the right way, we may see more legal jobs and more legal work available,” Arruda said. “We will always need lawyers, it’s just what it means to be a lawyer will change, and part of that will be reliance on software that is enabled by artificial intelligence.”
Originally published at www.legaltechnews.com.
