Court Acceptance of Predictive Coding? Still a Waiting Game

Created on July 26, 2012

Director of Marketing Programs

The emergence of predictive coding to streamline document review is like any new technology that hits the market, it takes time for users to accept its functionality and validate whether the application of the technology match its promised functionality and benefits. As evidenced by the Da Silva Moore, et. al. v. Publicis Groupe, et. al., predictive coding has, in some courts, passed hurdle number one with respect to functionality, as well as the potential cost saving benefits. What still remains unanswered is whether the promise and potential cost savings of predictive coding can be applied as a proportional solution for all e-discovery cases.

Global Aerospace Inc, et. al. v. Landow Aviation, L.P. dba Dulles Jet Center, et. al. (Vir. Cir. Ct. April 23, 2012), the latest decision involving predictive coding addresses the issue of proportionality and what role this new technology can play during the discovery cycle. In this case, the Virginia circuit court issued a court order approving the use of predictive coding during discovery and became the first state court to endorse the use of predictive coding outside the federal realm.

Before discovery began, the defendant, Landow Aviation, proposed the use of predictive coding to “retrieve potentially relevant documents from a massive collection of electronically stored information," arguing that a manual review of more than 2 million documents would cost $2 million and only 60 percent of all potentially relevant documents would be identified. The plaintiffs, Global Aerospace Inc, et al, objected to this form of document review based on the uncertainty surrounding the accuracy of predictive coding during document review. To bolster their argument, the plaintiffs went on to note that no court had ever compelled a party to accept produced documents selected by a “predictive coding computer program."

Even though the use of predictive coding may be testing uncharted waters, the Virginia circuit court approved the use of predictive coding in a short, two-page court order stating “Defendants shall be allowed to proceed with the use of predictive coding for the purposes of the processing and production of electronically stored information."

While this court order does endorse and enable defendants to use predictive coding, there are still numerous unanswered questions that need to be addressed:

  • What parameters will be set by the parties on how predictive coding will be used?
  • Are there specific guidelines defendants must follow when using predictive coding?
  • Will predictive coding be used in additional to human review? If so, in what capacity?
  • What sampling techniques will be applied to ensure that production set includes all potentially relevant ESI?

Like in Da Silva Moore, defining and agreeing to a predictive coding protocol that outlines the process of how to apply the technology is key. The court does not expect perfection; just reasonableness that matches the proportionality and required results for the case at hand. It's up to the legal scholars and attorneys to work together to define the accepted protocols and create a template for how to utilize predictive coding in a manner that is defensible and accurate.


The case law surrounding predictive coding is still evolving, yet we expect many other courts in the United States to endorse the use the newest technology, like Virginia. Cases, like Global Aerospace, bring to light that courts are considering other avenues outside of human document review and keyword searching to find responsive information. With corporate ESI volumes growing exponentially, the only way to stay in compliance with Federal Rule of Civil Procedure (FRCP) 1 (the FRCP “should be construed and administered to secure the just, speedy, and inexpensive determination of every action and proceeding") is to look outside the box and identify to new technologies, like predictive coding which can bring greater proportionality and cost savings to the discovery process.

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.