default logo

The Impact of IBM’s Watson on eDiscovery

Last week, the New York Times published a great article, IBM to Announce More Powerful Watson via the Internet (13 Nov 2013).  Will the availability of Watson in the Internet at an affordable price eventually have an impact on eDiscovery and document review?

That question came to mind immediately because on the same day, I wrote a blog post asking whether law firms would begin adopting more “horizontal” software.   That is, with the advent of massive cloud systems used by millions, will legal market specific software (“vertical”) continue to maintain its grip.

I asked several of my eDiscovery expert friends the question about Watson and eDiscovery.  Specifically, I wrote:

“I wonder if this technology could, in the not distant future, supplement or even supplant dedicated computer-assisted review (predictive coding) eDiscovery tools. I would like to write a blog post that reports on the opinion of eDiscovery thought leaders. If you are willing to write a short answer on your thoughts about whether Watson on the Web will have an impact on discovery, I would publish it as a quote”

Here are there answers, captioned by expert-author.  All answers are direct quotes, published with permission, in the order in which I received answers.

 

Jeff Brandt, Legal Technology & KM Consultant

It will happen.  Watson will impact eDiscovery and many other areas of law.  Over two years ago I wrote an article, The law firm of DLA, Watson, Siri & Wal-Mart. I set it in the year 2065 (primarily because I am a Thunderbirds TV Series fan) but I “predicted” that the first law firm Watson would come online in 2014 – next year. My timing may well be off.  It will probably take an ABS and outside investment to do (Can you say Wal-Mart?), but I am convinced it will happen. That will forever change lawyers’ roles and fundamentally change the law firm staffing model.

 

Jeff Fehrman, MindsEye

This is quite refreshing and will have an enormous impact on the use of data in solving real world problems. Yesterday [Nov 13], at a legal technology symposium a few individuals openly questioned the validity of applying technology like this in the discovery space. While the work Watson displayed in taking on Ken Jennings at Jeopardy was quite impressive; the impact on fields like healthcare, education, finance, and legal (eDiscovery) are where the power of technology has the chance to change entire industries. Some industries are likely to be negatively impacted by this type of technology, and many from these industries will resist change. What the people in these industries need to clearly understand is that with the pace of evolution in technology, change is going to happen and it has the potential to change quickly. Understanding data has everything to do with the process of discovery, advanced analytics and machine learning solutions like Watson will help people in realizing that data is an asset and the data holds many of the answers to questions that improve industries. The key is having the capabilities to ask the right questions, IBM has provided a giant leap forward in this respect.

 

Sherry Katz, Elluma Discovery

Thanks for sending this. It’s an interesting development – the irony is that I’ve been explaining analytics to lawyers for the last two years by referencing Watson and how it has been used in cancer diagnosis.  It seems logical that Watson could be tapped to mine ediscovery materials and for predictive coding.  A lot of my clients are medium sized law firms, not the big giants.  A computer forensics/ediscovery firm like ours can level the playing field with the big firms by giving smaller firms the technology to handle large amounts of data.  We are always looking to leverage technology available outside the legal marketplace, and tapping into Watson potentially could offer more democratization of the availablity of artificial intelligence tools.  We are probably still 15-20 years away from a market where lawyers have any broad understanding of how to make use of these tools.

 

David Carns, Legal Discovery, LLC

It is clear that distributed computing, machine learning and predictive analytics are driving much of the innovation in computing. Such technologies are making it possible to identify spam better, determine what people are likely to click on, or which documents may be responsive to a request in litigation. And while IBM’s Watson does not have an exclusive hold in this arena (think Google’s Predictive API or many others), the costs of using such techniques are being driven down at an exponential rate. It will be exciting to see what new developers are able to do with this technology as the costs of implementation continue to decrease.

 

Gregory P. Bufithis, Esq. Founder/CEO, Project Counsel

I am not the only one who started thinking about the potential e-discovery implications of IBM’s announcement last week.  Greg has written an in-depth piece, Watson as a service: IBM preps artificial intelligence in the cloud, and e-discovery asks “can we have a bite? over at his Project Counsel website.  Greg has spent a fair bit of time looking at and thinking about Watson.

With his permission, I am providing a teaser of some of his thoughts but you’ll have to read his whole piece to take in the impact on e-Discovery.

“The biggest news last week was probably the IBM announcement that it was offering Watson’s machine-learning system in the cloud, thereby expanding artificial intelligence’s (AI) frontiers and also monetizing the work as it happens. More importantly, IBM’s Watson announcement is significant, not necessarily because of the sophistication of the Watson technology, but because of IBM’s ability to successfully market the Watson concept….

The future of search is to not search at all, as Stephen Arnold of Beyond Search is fond of saying. And although this may sound contrarian, if you follow his columns and those of writers in KMWorld you know we are on the threshold of search technology that will eliminate the need to explicitly ask for information…

Its application to e-discovery and legal search is still to be determined … Watson pulls up stuff that a “mortal” cannot because it is looking for semantic links, not just doing text-matching based on keywords. … when Watson was rolled out for human resources departments … at a special human resources technology conference … attendees saw first-hand how cognitive computing systems that are taught not programmed provide personalized, evidence-based, contextually relevant responses to questions posed in natural language.”

  1. William Kellermann

    Reading all this makes me think of the scene in the file “The Hitchhiker’s Guide to the Universe” where the monolithic computer states the answer “42″ and the rest of the movie is devoted to discovering the correct question. Perhaps, licensing and royalties aside, the proper name for Watson is ‘Spock.’

    There are two issues with Watsonesque AI – the ability to ask the right question and the difficulty in truly measuring the accuracy of the answer. Yes we can measure the probablity of an answer’s accuracy but not with absolute certainty.

    If we had measurable knowledge of the contents, it would be easy to test Watson, but then we wouldn’t need it. What Jared Diamond has shown us is the true difficulty of applying science to human affairs (social science) which by its very nature cannot stand up to true scientific rigor, based on moral and ethical impediments.

    Analytics are a great tool. But Watson in any form is not the omnipotent ‘easy’ button.