This browser is not actively supported anymore. For the best passle experience, we strongly recommend you upgrade your browser.
Welcome to Reed Smith's viewpoints — timely commentary from our lawyers on topics relevant to your business and wider industry. Browse to see the latest news and subscribe to receive updates on topics that matter to you, directly to your mailbox.
| 1 minute read

A Better Way to Identify Potentially Privileged Documents

Advances in technology, such as predictive coding (aka “TAR”), have made it easier and more efficient to identify relevant documents for production and use in litigation matters. Even predictive coding, however, has not been very reliable in identifying documents subject to attorney client privilege or the work product doctrine. Privilege decisions can be dependent on roles and relationships rather than simply the words contained in the documents, which TAR analytics tend to focus on. Now, however, there are technological solutions more effective than simple keyword privilege screens for improving the efficiency of privilege identification. The leading example is the AI solution developed by Text IQ, a company that was acquired by Relativity earlier this year and is now being integrated with Relativity’s e-discovery platform.

In a head-to-head test against a privilege word screen (starting with the template included in the E-Discovery App), the software beat the privilege screen decisively at reducing both types of error: (1) identifying as potentially privileged documents that are not subject to privilege or work product protection, and (2) failing to identify all documents that qualify for privilege protection. In our test, run against hundreds of thousands of documents in a real litigation matter, Text IQ’s solution was able to filter out 26,000 more documents than the privilege word screen caught. Even more impressively (since we are most concerned with minimizing the risk of producing privileged documents), the Text IQ software found 99.9% of the privileged documents (as confirmed by later attorney review of all of the documents), including 50 that the traditional privilege screen missed. You can find a more detailed description of the test and results here.

The precise savings that can be achieved by using this technology depends on the workflow in any particular case, and the cost of applying the Text IQ solution, but it is fair to say that this technology can yield significant time and cost savings, as well as greater accuracy, in many cases and investigations.

Text IQ’s AI identified 99.9% of the documents ultimately withheld by Reed Smith as privileged.


e-discovery, tar, predictive coding, ai, artificial intelligence, ediscovery, attorney client privilege, privilege, document review, textiq