Judge Andrew Carter of the Southern District of New York recently adopted Magistrate Judge Andrew Peck's first-of-its-kind decision approving and encouraging the use of computer-assisted document review. Da Silva Moore v. Publicis Groupe, No. 1:11-cv-1279 (ALC) (AJP) (S.D.N.Y. Apr. 26, 2012) ("Carter decision").

In Da Silva Moore, a gender discrimination case, Judge Peck adopted a predictive coding protocol for the review of three million e-mails. No. 1:11-cv-1279 (ALC) (AJP) (S.D.N.Y. Feb. 24, 2012) ("Peck decision"). Computer-assisted review, or "predictive coding," refers to the use of a computer algorithm that "learns" to find relevant documents based on a human reviewer's classification of a document subset. This allows the legal team to review a smaller proportion of documents, and to focus on the documents that are most likely to be responsive. Computer-assisted review promises to reduce the number of reviewers and time to complete the review, thereby reducing discovery costs. Like manual review, a good predictive coding protocol includes quality control testing to confirm that the computer really "got it."  

Does it work? Not all predictive coding algorithms are created equal, but studies demonstrate that on average, computerized searches are at least as accurate – if not more so – than manual review. See Andrew Peck, Search, Forward, L. Tech. News, Oct. 2011 (citing Roitblatt, Kershaw, and & Oot, Document Categorization in Legal Electronic Discovery: Computer Classification vs. Manual Review, 61 J. Am. Soc. Inf. Sci. & Tech., no. 1, 2010 at 70-80 and Grossman & Cormack, Technology-Assisted Review in E-Discovery Can Be More Effective and More Efficient Than Exhaustive Manual Review, 27 Richmond J. Law & Tech., no. 3, 2011 at 1-48.)

Judge Peck, a proponent of computer-assisted document review, was intent on his decision paving the way for the use of this technology in litigation: "This judicial opinion now recognizes that computer-assisted review is an acceptable way to search for relevant [electronically stored information] in appropriate cases." Peck Decision at 2. "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." Id. at 25.  

Judge Carter adopted Judge Peck's rulings over plaintiffs' objections, finding that "they are well reasoned and … consider the potential advantages and pitfalls of the predictive coding software." Carter Decision at 3. "There simply is no review tool that guarantees perfection … [ and ] there are risks inherent in any method of reviewing electronic documents," Judge Carter wrote. "Manual review with keyword searches is costly, [ and ] though appropriate in certain situations … is prone to human error and marred with inconsistencies from the various attorneys' determination of whether a document is responsive." Id. at 4.  

With these precedents in hand, more courts are starting to weigh in on the use of computer-assisted review. A Virginia state court recently approved a defendants' request to use predictive coding over plaintiffs' objections. Global Aerospace Inc. v. Landow Aviation, Consolidated Case No. CL 61040 (Vir. Cir. Ct. Apr. 23, 2012) (noting that the approval was without prejudice to the plaintiffs' right to object to the continued use of predictive coding should issues arise). And, in an on-going case in the Northern District of Illinois, plaintiffs are asking that the defendants be ordered to redo their production using predictive coding. Kleen Prods., LLC v. Packaging Corp. of Am., No. 1:10-cv-05711 (N.D. IIl.). The court has yet to issue a ruling, but this case is being closely watched and should provide further insight into the future of predictive coding.