A Bright Future

Hayes Hunt and Jillian Thornton

Corporate Counsel

2012-10-01 00:00:00.0

A decade ago, document review meant a small militia of lawyers sitting in a windowless warehouse surrounded by bankers' boxes full of paper documents. Now, thanks to extreme information inflation, most document review takes place electronically. In order to keep up with the enormous volume of electronically stored information (ESI), lawyers have employed a method featuring a combination of keyword searches and manual review. Most importantly, e-discovery can be responsible for 70–90 percent of the client's cost of litigation. However, recently the universe of ESI has expanded in exponential fashion. Exabytes have devoured the smaller gigabytes. So what's the new flavor of e-discovery? Predictive coding.

Predictive coding is being used to run algorithms that allow for computer characterization of a massive set of electronic data for a fraction of the cost of more traditional methods (i.e., a cadre of lawyers). Case law is now catching up to the technology, and various judges are giving the green light for lawyers to employ predictive coding in e-discovery without running afoul of the rules.

The proper use of predictive coding, especially in large-data-volume cases, provides huge benefits: Predictive coding of ESI takes much less time, saves a lot of money, and is often as accurate or more accurate than manual review. Of course, predictive coding also can be problematic if, for example, privileged documents are disclosed.

A recent study by Rand Corp., which includes 57 case studies from eight large corporations, shows that the cost of e-discovery can be grouped into three main categories: collection, processing, and review. Amazingly, the review phase accounted for 73 percent of the costs incurred during e-discovery. Predictive coding works to drastically reduce the number of documents that are manually reviewed by lawyers.

Here's how it works: The first step in the process is that lawyers review a small sample of documents and code those documents for relevance or privilege or subject matter. The software then studies the sample set and applies the coding principles that it has learned to a larger set of documents. Then, the lawyers review the computer-coded documents to further teach the program how to code. This program continues until the software identifies only relevant documents. After coding is finished, the software can be used to select a small, random population of documents for lawyers to perform quality-control checks. If errors are found, the lawyers code more sample documents until accuracy of the coding reaches an acceptable level. Then the review is complete. The software can reduce the documents that need to be manually reviewed from a set of 2 million, for example, to only 3,000–5,000 documents.

The difficulty with implementing processes such as predictive coding is that the technology is so new that these methods are fairly untested in court. However, the first wave of cases discussing the propriety of predictive coding has illustrated that it is indeed going to be accepted as an appropriate discovery tool. Earlier this year, in a landmark decision, U.S. Magistrate Judge Andrew Peck of the Southern District of New York authorized the use of predictive coding in Da Silva Moore v. Publicis Groupe . "What the bar should take away from this opinion is that computer-assisted review is an available tool and should be seriously considered for use in large-data-volume cases where it may save the producing party (or both parties) significant amounts of legal fees in document review," the judge wrote.

The district court ultimately adopted Peck's evidentiary rulings in that case. In his opinion, Peck was careful to point out that the plaintiffs consented to the defendant's use of predictive coding, and the discovery dispute merely concerned the implementation of its use. Thus, lawyers were left to wonder what would happen if the parties did not agree to the use of predictive coding.

The same week that Peck's ruling was affirmed, a state court judge in Virginia approved the use of predictive coding in a case over the objections of the opposing party. In Global Aerospace v. Landow Aviation , Judge James Chamblin ordered that defendants could use predictive coding, despite the plaintiff's objections that the technology was not as effective as manual review. Chamblin disagreed and ordered the predictive coding for the production of the defendant's ESI, provided that the receiving party would still have an opportunity to question the completeness of the contents of the production or the ongoing use of predictive coding.

Predictive coding and other automated methods of e-discovery obviously have limitations. Peck, in Da Silva Moore , emphasized that his approval of predictive coding was not universal: He did not order the use of predictive coding, he stated that computer-assisted review is not required in all cases, and he did not endorse any particular vendor or predictive coding tool. In addition, it is the responsibility of the lawyers to understand the predictive coding program and how it works so that they can demonstrate the method's reasonableness if it is called into question. Just as with the traditional e-discovery tool of keyword searching, lawyers must engage in careful planning and sufficient quality control to ensure the accuracy of the program. Finally, lawyers should cooperate with opposing counsel and be transparent in their use and the scope of predictive coding in order to avoid unnecessary discovery disputes.

With regard to privilege review, lawyers utilizing predictive coding need to be especially vigilant not to inadvertently produce privileged documents. Although predictive coding can be used to assess privilege as well as relevance, lawyers need to evaluate the benefit compared to the risk of disclosure. Under the federal rules governing clawback, a "disclosure of a communication or information covered by the attorney-client privilege or work production protection . . . does not operate as a waiver in a federal or state proceeding if (1) the disclosure is inadvertent; (2) the holder of the privilege took reasonable steps to prevent disclosure," as well as reasonable steps to correct the error.

Thus, whatever method of ESI review lawyers use, it must rely on reasonable steps to prevent disclosure. This is a subjective standard, and lawyers using predictive coding would be wise to carefully document the process of how they code for privilege.

Given the ever-expanding universe of ESI, most lawyers would be wise to consider using computer-assisted review and especially predictive coding. After all, the research has shown that predictive coding is more precise, makes fewer errors, and identifies more relevant documents than human reviewers. When you add millions of pages of documents, fatigue plays a role for human reviewers. On the basis of these factors, in addition to the dramatic saving of time and money, it is clear that predictive coding and similar methods are going to revolutionize how we conduct e-discovery.

Hayes Hunt is a member of Cozen O'Connor in the firm's commercial litigation and criminal defense and government investigations practice groups. Jillian R. Thornton is an associate in the firm's commercial litigation group. A version of the article appeared in sibling publication Law Technology News.