Back in January we noted that — ready or not— 2012 will be remembered as the year that predictive coding finally arrived. As we approach midyear, all signs point to that prediction coming true. A couple of prime examples:
- Earlier today The Wall Street Journal ran an article entitled Why Hire a Lawyer? Computers Are Cheaper.
- On Friday, Judge Peck entered a 56-page order in the Da Silva case refusing to recuse himself from a matter in which the plaintiff accused the magistrate of having too cozy a relationship with the predictive-coding industry and its advocates.
What does this tell us?
As we “predicted,” the discussion around automated review has turned from theoretical to practical. Now the heavy lifting begins.
And it’s not simply the advanced clients looking to save money who are considering automated review strategies. To the contrary, in the Kleen matter (which is still pending, or should I say even more ominously, looming), the producing parties want nothing to do with predictive coding. They simply want to do their discovery the way they’ve been doing it for the past decade — interviewing custodians, conducting keyword searches, and having attorneys execute document review projects.
To their credit, the defendants have gone one step further and performed some statistical sampling to attempt to validate the efficacy of their search strategies. However, the plaintiffs in the Kleen matter say that the old-fashioned way simply isn’t good enough. They want the defendants to use predictive-coding tools across their data set to ensure that relevant data is not being left behind.
The irony is that two of the cases in today’s news — Kleen and Da Silva — both have predictive coding at the center of their discovery disputes, but the positions taken by the producing parties are 180 degrees apart. One producing party (Da Silva) wants to use automated review to save money and the defendants want none of it, fearing that predictive coding will produce inferior results; the plaintiffs in in the other matter (Kleen) have pleaded for the use of predictive coding to ensure that the production set is complete, and the defendants have objected, in part, because predictive coding will increase their cost. Two cases, nearly identical issues, and opposite positions taken by the respective parties.
While no one knows how these cases will turn out (and that goes for the Global Aerospace Inc., et al, v. Landow Aviation, L.P. dba Dulles Jet Center, et al, mentioned in today’s Wall Street Journal article), it’s clearly time for clients, counsel, providers and courts to get in the game — and to make some rules along the way.