Magistrate Judge Peck of the Southern District of New York considered at length the use of computer-assisted document review software and approved a discovery plan utilizing such predictive coding technology over several objections by the plaintiff.  Moore v. Publicis Groupe S.A., No. 11-CIV-1279 (S.D.N.Y. Feb. 24, 2012). The case concerned a claim of company-wide gender discrimination and required the review of over 3 million emails.  The defendant proposed to use computer-assisted review to narrow the field of documents for production.  As the court explained, computer-assisted review utilizes software “that use[s] sophisticated algorithms to enable the computer to determine relevance, based on interaction with (i.e., training by) a human reviewer.”  In essence, attorneys review a small sample set of the documents at issue, grade the documents for relevancy, and then allow the software to grade the remaining field of documents using an algorithm that “learns” from the sample set.  The court approved the computer-assisted protocol, rejecting arguments that the predictive coding would be unreliable or would violate the Federal Rules of Civil Procedure.  The opinion generally advocates for a greater use of such software and attaches a detailed form protocol for adaptation in other cases.