We’ve all heard the threat: Robots powered by artificial intelligence soon will replace attorneys in e-discovery. That was even the suggestion of the 2011 headline in The New York Times, Armies of Expensive Lawyers, Replaced by Cheaper Software.
Not surprisingly, the actual Times article was less alarmist than the headline, making the point that AI and machine-learning tools can be used to lighten the load on lawyers, but not to take it away altogether. And the Times stepped back even further in a subsequent article, The End of Lawyers? Not So Fast.
“In recent months,” the second article reported, “a growing array of new studies have indicated that the relationship between technological advances and job displacement is more complex and nuanced than pessimists have suggested.”
For lawyers and businesses involved in e-discovery, this stuff matters because the e-discovery field was effectively ground zero for today’s exploding use of AI in law. Many lawyers’ first encounter with machine-learning technology was through technology-assisted review or predictive coding, and those technologies marked the first significant use of AI in any field of law.
AI’s Rise in Law
All of sudden, it seems, AI is everywhere in law. It is being used for due-diligence review, legal research, contract analysis and docket analytics for predicting litigation outcomes. Those uses are not necessarily outgrowths of AI’s use in e-discovery. After all, AI is beginning to permeate all aspects of technology and all sorts of industries. But AI in e-discovery probably gave those uses a confidence boost.
Within the legal industry, the ever-widening use of AI is chronicled in a new book, which also offers predictions for its future. In Robots in Law: How Artificial Intelligence is Transforming Legal Services, legal technology journalist Joanna Goodman surveys the development and use of AI in law and considers what it means for providers of legal services.
A key point she and the experts she quotes agree on is that AI is not going to replace lawyers, but rather enhance their efficiency and productivity. At the same time, AI is going to change the value proposition for lawyers and their clients, and those who adapt soonest will benefit most.
“Progress so far,” Goodman writes, “indicates that the current application of legal AI could well be part of an evolution that will be a game changer for legal services, not because it will change the basic premise of what lawyers do – or replace them all – but because it will create shifts in the value chain, and therefore change the legal business model in terms of legal services procurement, billing – and margins.”
The Value Chain for E-Discovery
If e-discovery was indeed ground zero for AI in law, it was also the proof of concept for those “shifts in the value chain” Goodman mentions. TAR offers many advantages in e-discovery, but the reason it took off as it did was ultimately economic – it was far cheaper for the client and far more efficient for the lawyers to use TAR to review large quantities of documents for e-discovery.
The turning point – or perhaps springboard – for TAR’s adoption can be traced to 2011, when two e-discovery researchers, Maura R. Grossman, then a practicing lawyer and now a research professor at the University of Waterloo, and Gordon V. Cormack, co-director of the Information Retrieval Group at the University of Waterloo, analyzed data from the 2009 TREC Legal Track involving the use of TAR processes. They concluded that TAR was not only more effective than human review at finding relevant documents, but also much cheaper.
“Overall, the myth that exhaustive manual review is the most effective—and therefore the most defensible—approach to document review is strongly refuted,” they wrote in the Richmond Journal of Law and Technology. “Technology-assisted review can (and does) yield more accurate results than exhaustive manual review, with much lower effort.” Their study found that TAR produced a 50-fold savings in cost over manual review.
It was not long after that when U.S. Magistrate Judge Andrew J. Peck issued the first-ever judicial opinion approving the use of TAR, Da Silva Moore v. Publicis Groupe. Not only did Judge Peck approve TAR for the case at hand, but he also broadly endorsed it as superior to other methods and urged its use. “Computer-assisted review appears to be better than the available alternatives, and thus should be used in appropriate cases,” he wrote.
AI in E-Discovery Today
So e-discovery was the first segment of the legal profession to adopt machine-learning technology and substantively incorporate it into day-to-day practice. But that wasn’t the end of the story.
While TAR’s initial use was for review of documents prior to production in litigation, its use has expanded significantly since then. It is now more commonly used for purposes such as privilege review, searching for “hot” documents in an opposing party’s production, and discovering the “stories” within collections of documents. (See, TAR 2.0 Capabilities Allow Use in Even More E-Discovery Tasks.) Some companies are also using it to speed compliance investigations.
Most recently, Catalyst has broadly incorporated machine learning and AI technology in its new Insight Enterprise platform for corporate e-discovery. Insight Enterprise uses a centralized repository to allow corporate counsel to better manage across all their legal matters. As new matters arise, documents can be promoted from the core repository to individual matter sites.
Machine learning is used in Insight Enterprise in two primary ways. One is for traditional review. Within individual matters, a company’s legal teams are able to use TAR in the traditional way of ranking documents for review and reducing the overall number of documents to review.
But machine learning can also be used across the Insight Enterprise platform. This means that corporate counsel and their outside legal teams can use TAR and machine-learning tools for investigations, early case assessment and for identifying documents to promote for individual cases. This makes the efficiency and time-savings advantages of TAR available across a broader range of use cases.
The fact of the matter is that AI is not a nefarious robot out for our jobs. Rather, it is a tool for helping lawyers and their clients achieve greater efficiency, productivity and cost savings. E-discovery was the first legal practice area to make widespread use of machine learning technologies, and today it continues to drive innovations in AI – not to replace lawyers, but to enhance their efficiency and effectiveness.