Exterro's E-Discovery Breakdown

The world of E-Discovery is constantly changing – let us break it down for you with a weekly dose of News, Resources, Case Law, and Humor, all written in a concise and easy to understand format.

< BACK TO ALL STORIES

The Basics of E-Discovery: Predictive Coding and AI

Created on September 14, 2018


Solutions Consultant at Exterro

What Is Predictive Coding?

Predictive coding is a technology that assists document reviewers during the review phase of the EDRM. Typically it uses machine learning to label documents after the reviewers have completed some review, creating a “seed set” of coded (or labelled) documents. It improves the efficiency of human reviewers by understanding how they’ve labelled documents and then using those criteria to apply labels to documents that have not been reviewed yet.

How Does Predictive Coding Differ from Artificial Intelligence (AI)?

Artificial intelligence is an overarching term where the technology is doing more than just helping you with the review phase. It might help you with other parts of your discovery. iPredictive coding is specific to review.

What Do the Courts Think of Predictive Coding and AI in E-Discovery?

In general, the courts have been receptive to the use of predictive coding and artificial intelligence in e-discovery. Some significant rulings from Hon. Andrew Peck, in rulings like Da Silva Moore v. Publicis Groupe and Rio Tinto PLC v. Vale SA, approved (but did not mandate) the use of technology assisted review (and predictive coding) during e-discovery review.

That said, attorneys have been slower to adopt predictive coding and artificial intelligence into the e-discovery process. And when predictive coding or AI are used in e-discovery, the courts have reinforced the importance of both parties being transparent and cooperative to avoid turning these technologies into yet another grounds for e-discovery disputes.

What New Predictive Coding and AI Technologies Are on the Horizon?

The future of predictive coding and artificial intelligence and e-discovery is very interesting and rapidly evolving. The rise of deep learning techniques in artificial intelligence is already filtering into e-discovery technology. Using deep learning techniques, the e-discovery technology will increasingly recommend specific steps or tasks for the e-discovery professionals to perform, especially during review. In short, both of these areas are undergoing transformations to assist the goals of making e-discovery both more efficient and more accurate.

If you’re interested in learning more about predictive coding and artificial intelligence in e-discovery, download our recent infographic titled What AI Is—And Isn’t—In E-Discovery.

Of course, Exterro’s Basics of E-Discovery guide dives deeper on these and other topics and is available as a downloadable PDF as well.