By Rich Smith
Technology assisted review (predictive coding) for ediscovery still strikes fear in the hearts of many lawyers. Indeed, in the last ABA Legal Technology Survey (2018) only 12% of the respondents reported using predictive coding to process or review e-discovery materials. Many reported significant concerns about cost and the validity of the process.
In the following Guest Post, Rich Smith discusses the use and advantages of technolgy assisted review and dispels many of the concerns about use.
Rich Smith is a Senior eDiscovery Consultant for Page One Legal. With 7 years of legal technology experience, Rich will be happy to discuss your current technology needs and help craft a vision to be more successful with your practice. He can be reached at: firstname.lastname@example.org or 502.376.5829. Page One Legal handles all aspects of eDiscovery and Digital Forensics. With 13 years experience, they provide eDiscovery platforms with both Relativity and Relativity One.”
I would like to thank Rich for his willingness to publish his article on this site. I welcome guest post submissions from responsible authors on topics of interest to this site’s readers. Please contact me directly if you would like to submit a guest post. Here is Rich’s guest post.
Lawyers fear technology.
However, SUCCESSFUL lawyers are EMBRACING technology.
THE MOST SUCCESSFUL lawyers are utilizing technology to assist them, especially during review of large amounts of documents.
Wait…technology…assisted…review? T.A.R.? Hmm, isn’t that sometimes called, “predictive coding”?
Predictive coding is the automation of document review. This typically works by taking information gained from manual coding and automating that logic to a larger group of documents. Reviewers use a set of documents to identify potentially responsive documents and then train the computer to identify similar ones.
Basically, the machine is learning and studying you during review so it can suggest likely relevant documents. It’s a form of supervised learning.
What’s supervised learning? It’s the machine-learning task of learning a function that maps an input to an output based on example input-output pairs. It infers a function from labeled training data consisting of a set of training examples. In supervised learning, each example is a pair consisting of an input object (typically a vector) and a desired output value (sometimes called the supervisory signal). A supervised learning algorithm analyzes the training data and produces an inferred function, which can be used for mapping new examples. This allows the algorithm to correctly determine the class labels and unseen instances.
You are already using Predictive Coding in everyday life.
GUESS WHAT: you are already using Predictive Coding in everyday life.
Pandora. You select a station and give the ole thumbs up or thumbs down to each song that comes across the platform. The algorithm learns your taste of tuneage and suggests plenty of songs you will enjoy.
Amazon. Ever notice what happens AFTER you order something online? Wow, check out all the speakers that go well with that sweet big-screen tv you just bought.
Facebook. Well, Facebook is always watching and learning from you anyway. (Allegedly).
Ok, so if we are using T.A.R. in real life, why aren’t we using it for our caseload? Partly because not every case warrants it…but partly because of FEAR.
Fear of cost. Fear of errors. Fear of complexity. Also, the fear of it not being allowed in courts, right?
Back in February of 2012, renowned New York Magistrate Judge Andrew Peck made a ground-breaking ruling with Da Silva Moore v. Publicis Groupe et al (S.D.N.Y. 2012) Court Description: OPINION AND ORDER: There simply is no review tool that guarantees perfection. The parties and Judge Peck have acknowledged that there are risks inherent in any method of reviewing electronic documents. Manual review with keyword searches is costly, though appropriate in certain situations. However, even if all parties here were willing to entertain the notion of manually reviewing the documents, such review is prone to human error and marred with inconsistencies from the various attorneys deter mination of whether a document is responsive. Judge Peck concluded that under the circumstances of this particular case, the use of the predictive coding software as specified in the ESI protocol is more appropriate than keyword searching. The Court does not find a basis to hold that his conclusion is clearly erroneous or contrary to law. Thus, Judge Peck’s orders are adopted and Plaintiffs’ objections are denied.
(Signed by Judge Andrew L. Carter, Jr on 4/25/2012)
“This judicial opinion now recognizes that computer-assisted review (i.e., T.A.R.) is an acceptable way to search for relevant ESI in appropriate cases.”
In 2015, Judge Peck went on to state that “In the three years since Da Silva Moore, the case law has developed to the point that it is now black letter law that where the producing party wants to utilize TAR for document review, courts will permit it.”
Ok, we know it’s allowed in court. Why else aren’t we using T.A.R.?
Any good Litigation Support or vendor worth their salt will prove the efficiency of T.A.R. This will OVERWHELMINGLY cut review cost for the client
*Sounds expensive and my client won’t want to pay for it.
*Sounds really complex and some of our team isn’t that, “tech-saavy”.
*What if I train the machine incorrectly and we miss documents? How can we protect against overlooking key relevant documents?
Truth is, any good Litigation Support or vendor worth their salt will prove the efficiency of T.A.R. This will OVERWHELMINGLY cut review cost for the client. And, as the technology improves, so does the ease of use.
Point is, don’t be scared to apply A.I. to your matters. Skynet won’t be sending Terminators for at least 20 more years. So until then, start utilizing predictive coding and other tools to keep your clients happy.