Slaw: Complexity of Legal Writing and Analysis Still Stumps AI

Sarah Sutherland over at

Start-ups are pivoting away from legal analysis to subject areas that have more accessible datasets and less complicated source material, and those that haven’t frequently struggle to answer simple questions. There are many applications for automated analysis of legal documents, but as far as I can tell so far they tend toward extracting particular information such as judges’ names…”


Legal documents are some of the most complex writing in English, and it is unlikely that the nuance of what they mean will be an easy target.”

Covering Tax Reform (and Patents)….Oh My!

There is nothing quite like returning from vacation to a once-in-31-years, nearly 500-page tax bill needing quick attention (and needing a neat, tidy incorporation into my existing tax publication totaling over 2000 print and digital pages). If that isn’t sufficient, patents, first-to-file, the AIA, and the subsequent case law are thrown in for good measure here in 2018.

A “Happy New Year” to all.

Language Generator Mimics Fifty Shades’ Dummy Prose: Is There a Lesson Here for Legal Publishing?

Powering this software “generator” appears to require more human effort and input than People’s headline lets on though it is nonetheless amusing. The text result reads something like remedial freshman poetry (perhaps the inexact nature of the program’s vocabulary inadvertently mimics the abstractions and allusions of amateurish verse).

I do wonder what the corporate publishing world is making of this. If the executives indeed believe that words are cheaper and less meaningful than pennies, the answer may be painfully obvious.

Stanford’s CodeX Reveals the Many Possibilities for New Legal Technology

Stanford Law School’s CodeX Project is seeking fellows for the next academic year starting this fall.

CodeX, a joint endeavor between technology-field departments such as engineering and computer science and the law school at The Stanford Center for Legal Informatics, is where “researchers and entrepreneurs design technologies for a better legal system. CodeX’s broad mission, under Executive Director Roland Vogl, is to create legal technologies that empower all parties in our legal system and not solely the legal profession.” The advancements are intended to lead us toward the “next frontier” of legal technology.

A review of the projects featured on the Center’s website reveals some interesting Center goals including a visual representation of the players involved in the legal and political system. A tool such as this should be, in my humble opinion, unrelenting in its real world explication of these connections without protective coating intended to gloss over the ragged edges of the system. Ideally, it would serve as a monument of legal realism against the formalistic cant that composes much of the structure of legal information, legal institutions, and legal writing.

I could imagine it being (perhaps in a form not exactly the one envisioned by CodeX) an unintentional tutorial on public choice theory and the workings of mass democracy. The influence of lobbyists, special interest groups, corporations, and voters upon the regulatory, legislative and judicial processes would be revealed. Further, the differing goals, biases, and assumptions of common law courts versus those of administrative agencies and special statutory courts would be detailed and explored. It will be quite fascinating to watch and see what these projects actually lead to.