Regulators in the U.S. and abroad have been quick to recognize the value of technology-assisted review (TAR) workflows as important tools that can prevent massive data dumps and allow them to focus their analysis on critical and relevant data. A significant aspect of our experience producing data to government agencies has developed in antitrust matters relating to mergers and acquisitions. When the Government issues a request for additional information (referred to in the U.S. as a Second Request), parties have a limited time frame in which to collect, process, analyze/review, and produce a potentially large set of data, collected from an ever-expanding number of sources. In these circumstances, it would be challenging (if not impossible) to reach a point of substantial compliance without the use of tools that can quickly and defensibly identify potentially responsive data. Although some counsel may still favor the use of search terms to identify responsive content, the clear trend for complying with Second Requests in the U.S. is to deploy a TAR workflow.

The Antitrust Division of the Department of Justice (DOJ) requires counsel to provide information at the inception of the project regarding whether the party intends to utilize TAR as part of the analysis process. Counsel must provide such information as:

1. the identity of the eDiscovery provider,
2. a description of the TAR process (including the TAR software deployed),
3. the lawyers who will be training the algorithm,
4. targeted metrics for recall and margin of error, and
5. the type of data that will be run through the TAR engine and identification of data that will not be included in the TAR process.

Given the need to quickly assess responsive content and to get it analyzed/reviewed in an expeditious matter, Standard Active Learning protocols (TAR 1.0) have become the accepted norm. Once the DOJ has signed off on the TAR Protocol, it then typically expects to receive and review several randomized sample sets of documents that fall below the TAR cut off score in order to validate the results of the TAR process. One significant benefit of the DOJ’s approach is that the parties have a clear understanding, before they begin, about the details of the TAR process and how the results will be evaluated. Historically, the Federal Trade Commission (“FTC”) has not required the same level of up-front negotiation and scrutiny into a party’s proposed TAR methodology/process. However, in a recent blog post, the FTC indicated that it is moving to an approach that is more in line with the DOJ, and has begun to ask for a TAR Protocol at the inception of the matter. Making the Second Request Process Both More Streamlined and More Rigorous During this Unprecedented Merger Wave, https://www.ftc.gov/news-events/blogs/competition-matters/2021/09/making-second-request-process-both-more-streamlined?utm_source=govdelivery (Sept. 28, 2021).

Although these approaches are not necessarily indicative of how other government regulators will respond to the use of TAR when receiving data and what they will require, there are certain general points to bear in mind when producing to a government regulator or legislative body:

• Be sure to review and understand any operative ESI Protocol that the relevant agency or legislative body may have promulgated. Apart from the use of TAR, it is vitally important that legal counsel understands how to produce data to that agency. Any government requirements may also have an impact on how the document review process will take shape.
  • As an example, if an agency prohibits any “downgrades” of documents from responsive to non-responsive after the completion of the TAR process and the identification of a potentially responsive set, the best approach may be only to review the documents flagged as potentially privileged and those that were not run through the TAR engine.
  • This workflow has also become common for U.S. Second Requests. In that case, to better understand what is in the produced data, the responding party can perform analysis of the data set to be produced (after the TAR 1.0 process has been completed) using a Continuous Active Learning (TAR 2.0) workflow.
• Determine whether the agency requesting the production of data has requirements around the use of TAR. Even if not specifically required or set out in a policy document, it is always best to engage in these discussions with the government attorney who is assigned to your matter early in the process so that he or she has a clear understanding of how you wish to proceed. If the legal team is not fully familiar with the details and nuances around the use of TAR for productions to the government, they should feel comfortable including the outsourced eDiscovery provider on these calls or, at a minimum, having it provide guidance to counsel beforehand.
• The legal team should work closely with their eDiscovery provider to oversee the TAR process. The results of using TAR are always enhanced when there is a close working relationship between counsel and the eDiscovery provider.
• Be sure to follow any reporting requirements that the agency may require.

This blog post is an excerpt from the Chapter titled “Outsourced Document Review: Data Intelligence, Technologist Lawyers, Advocacy Support” by Edward Burke and Allison Dunham, which appears in the Thomson Reuters treatise eDiscovery for Corporate Counsel (2022). Reprinted with permission, © 2022, Thomson Reuters. Brett Beeman also contributed to this blog. 

A link to the book appears below: