When you think of judicial acceptance of predictive coding technology in document review, what probably comes to mind is Judge Andrew Peck's 2012 ruling in the highly publicized Da Silva Moore v. Publicis Groupe case. Da Silva Moore was the first case that formally accepted the use of predictive coding within discovery. The district court who affirmed Judge Peck's ruling, justified its ruling based on finding no evidence to prove that using predictive coding is less appropriate than using traditional techniques (i.e. keyword searching) to identify responsive information in the document review stage.
Unfortunately the parties in Da Silva Moore could only initially agree to use predictive coding but could not agree on the specific terms and conditions on how to employ the technology. Considerable delays and questions have arisen, making some wonder if predictive coding will even be used in this case at all. So as important as Da Silva Moore was, In re: Actos (Pioglitazone) Product Liability Litigation (W.D. La July 27, 2012), may eventually become the bellwether case on how to implement and effectively utilize the predictive coding technology. In Actos, the court accepted the general use of predictive coding, and, unlike Da Silva Moore, all parties were able to come to agreement on mutually accepted terms and conditions surrounding the actual use of predictive coding technology during document review.
In coming to this agreement all parties decided proactively before discovery to cooperate and collaborate in formulating detailed steps on how to effectively and fairly incorporate predictive coding within the document review process. Highlighted below are a few of the general predictive coding conditions found in the ESI protocol:
- Defined specific conditions for formulating a sample collection population to train the predictive coding software
- Assigned experts to understand the specifics of the coding process behind the software
- Established a formalized training process for specified individuals who would be using the predictive coding software
Aside from these conditions, all parties did agree via document sampling, to retain the right to review documents for relevancy and privilege after predictive coding has taken place, prior to production.
THE E-DISCOVERY BEAT'S TAKE
In 2012 at the federal level, as evidenced by Actos and Da Silva Moore, the use of predictive coding is being established as an acceptable technique in the courts to identify responsive documents during the document review stage. But this trend of accepting predictive coding seems to be predicated upon the two parties cooperating and agreeing to a protocol on how to apply this new technology. Based on Actos and Da Silva Moore parties in most scenarios will demand transparency into how the predictive technology works (i.e. related to statistical sampling protocols) so that they can clearly understand what will be produced and in turn received with a reasonable level of confidence.
Beyond just utilizing predictive coding during document review it leads one to contemplate if legal teams can begin to leverage similar machine learning technology like predictive coding to reduce costs and streamline e-discovery activities before the review stage (pre-collection). Predictive technologies delivers advanced machine intelligence to automate the identification and classification of ESI across the EDRM, from pre-collection through review. Whereas predictive coding is applied only after an ESI collection during the review phase, the advancement of predictive technologies enables legal teams to formulate case strategy in advance of collection during early case assessment, at the point of collection and throughout review. This offers the potential to not only streamline review and e-discovery costs but also deliver greater intelligence across the full spectrum of legal matters, including litigation, Rule 45 subpoenas, second requests, regulatory compliance and internal investigations.
To learn more about predictive technologies and how they can help legal teams narrow the ESI funnel early in the discovery process, download Exterro's webcast, “Leveraging Predictive Technologies across the EDRM" here.
Mike Hamilton, J.D. is a Sr. E-Discovery Analyst at Exterro, Inc., focusing on educating Exterro customers, prospects and industry experts on how to solve e-discovery issues proactively with technology. His e-discovery knowledge, legal acumen and practical experience give him a valuable perspective on bridging the gap between IT and legal teams. You can find him on Google+, Twitter and Linkedin.