It has been a busy week in the e-discovery world. On Wednesday, a county court in Virginia ordered litigants to use predictive coding, despite the plaintiff’s objection. Last week, the plaintiffs in Da Silva Moore v. Publicis Groupe et al. tried to boot Magistrate Judge Andrew Peck from the case, as well as roll back his landmark ruling, which endorsed the technology for the first time.
Well, despite the haters, no one can stop the march of progress. A federal judge weighed in on Da Silva Moore yesterday. It looks like the score is Robots 1, Old-school Attorneys 0….
From Law Technology News:
Magistrate Judge Andrew Peck’s February ruling allowing the use of predictive coding, aka computer-assisted review, to weed through millions of electronic documents in discovery has been upheld by a federal judge.
“There simply is no review tool that guarantees perfection,” said Judge Andrew Carter of the U.S. District Court for the Southern District of New York, who turned aside several objections to Peck’s ground-breaking decision adopting an electronically stored information protocol in the gender discrimination case of Monique da Silva Moore, et al. v. Publicis Group SA, et al, 11 Civ. 1279.
“Manual review with keyword searches is costly, though appropriate in certain situations,” Carter said. “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’ determination of whether a document is responsive.”
If you are interested in seeing Judge Carter’s full ruling, here is the PDF.
Of course, you all know the real news here, right? Skynet will be going live shortly. Its first targets will be… document review attorneys whose jobs it wants to steal.
In all seriousness, according to LTN, here is how Judge Peck justified recommending the technology (it’s worth noting that Da Silva involves discovery of some three million emails):
Intended as an alternative to manual review of mountains of documents by junior law firm staff, the method involves reviewing and coding a “seed set” of documents by a senior lawyer or a small team at the firm. The computer then identifies properties of those documents that it then uses to code other documents.
“As the senior reviewer continues to code more sample documents, the computer predicts the reviewer’s coding (or the computer codes some documents and asks the senior reviewer for feedback,)” Peck said. “When the system’s predictions and the reviewer’s coding sufficiently coincide, the system has learned to make confident predictions for the remaining documents.”
Peck said the senior lawyer would usually have to review only a few thousand documents to effectively train the computer.
In Moore, Peck’s protocol calls for using a 95 percent “confidence level” to create a random sample of the email collection, with a sample of 2,399 documents to determine relevant documents for the seed set that will be used to train the predictive software.
In all fairness, it seems reasonable that litigants would be a little unsettled about this new technology. Because A) it’s new and scary. Most people — even nonlawyers — get intimidated by new, untested technology. In more positive terms: it takes guts to be an early adopter of anything. B) In crass terms, predictive coding means that you are handing over discovery to a computer. That’s an intimidating concept. Even if human document review is expensive, skull numbing, and not necessarily effective, it’s certainly more satisfying to yell at a young attorney who screws up then to yell at Watson.
Even if predictive coding really is the way of the future, it will probably be a while before we stop seeing intense fights in nearly every case that seeks to employ it. But then again, intense discovery fights already happen all the time in all sorts of cases, so maybe this is just life as usual.
Peck Decision on Use of Predictive Coding Upheld in N.Y. Federal Court [Law Technology News]