Are document reviewers about to be replaced?

The electronic discovery community is all abuzz about U.S. Magistrate Judge Andrew J. Peck’s Feb. 24 opinion in which he recognized that “computer-assisted review is an acceptable way to search for relevant [Electronically Stored Information] in appropriate cases.” Peck, of the Southern District of New York, is one of several judges who has taken an active role in the debate on electronic discovery issues, along with Magistrate Judge John M. Facciola in Washington, D.C. and our own Chief Magistrate Judge Paul W. Grimm.

When Peck ruled from the bench Feb. 14, numerous blogs and websites and even the American Bar Association were quick to call the “decision . . . to require a so-called predictive coding protocol for automated e-discovery in an employment discrimination class action” one that could put document review teams’ jobs at risk.

In the current job market, an opinion that signals the death knell of any position — even document review positions — is highly concerning, particularly to young attorneys and law students.

But, there is good news. When Judge Peck issued his formal opinion, he clarified:

To correct the many blogs about this case, initiated by a press release from plaintiffs’ vendor—the Court did not order the parties to use predictive coding. The parties had agreed to defendants’ use of it, but had disputes over the scope and implementation, which the Court ruled on, thus accepting the use of computer-assisted review in this lawsuit.

Vendors, calm down. And document reviewers, relax. No court has mandated that attorneys use predictive coding or computer-assisted review.

While computer- or technology-assisted review (TAR), predictive coding and auto classification are not going to rid the world of all document reviewers anytime in the near future, it does hold a lot of promise. For those who are unfamiliar with the technology, in simple terms, TAR is a developing technology that allows computers to review and analyze a sample of documents that have been reviewed and tagged by human reviewers and then develop a complex set of terms and criteria by which the computers can “predict” which of the remaining documents contain similar and related relevant terms and characteristics.

Use of TAR technology will not wholly eliminate human reviewers from the document review process. Reviewers must still review and code, in detail, a sample set of data — usually 10,000-to-20,000 documents — and must monitor the computers’ relevancy (and irrelevancy) decisions and make any necessary adjustments. This means that TAR is a valuable tool only for larger cases — those for which the review would exceed the sample size.

Even for larger cases, however, many attorneys will be reluctant to adopt TAR-type technologies. TAR is new, and many will be uncomfortable with entrusting so much discretion to a computer. TAR may give some attorneys (and clients) sticker shock, and attorneys must be extensively trained on how to make the technology truly assist them.

That said, acceptance of TAR continues to grow, as evidenced by Peck’s opinion and the frenetic buzz at this year’s LegalTech New York conference.  At the same time, review platform vendors continue to refine the technology and are making versions increasingly more scalable to and feasible for smaller cases.

It is a development that should be investigated and understood, not feared.