One cannot open up their computer or turn on their television for any significant amount of time without seeing or hearing about artificial intelligence. The term evokes an almost immediate emotional reaction, often with ideas of a dystopian future where the human race is no longer master of the planet. Without delving too deep into that rabbit hole, I would instead leave The Terminator and other equally bleak futures out of this particular conversation and instead focus on artificial intelligence and the law.
The current application of artificial intelligence to the practice of law was a discussion topic at our most recent board of directors meeting for Loyola Law School. The discussion centered around the ability of a computer to perform a task or series of functions that had traditionally been the responsibility of a legal professional or team of professionals. After a lively discussion, I left with the three following questions:
• Are machines capable of what would be considered regular social interaction with a client such that the client would be comfortable interacting with the AI as their primary contact?
• What are the current applications of artificial intelligence in the law, or what can a machine actually do?
• What, if any, are the limitations to AI as it relates to the practice of law? Will computers continue to evolve, develop, learn and undertake increasingly more difficult and nuanced professional legal undertakings?
As a practicing lawyer, there are specific jobs I undertake on a daily basis that I have difficulty delegating to anyone, never mind a computer. Complexities inevitably arise with almost all legal matters and are often compounded by the clients and their expectations, assumptions and presumptions. How could one trust a computer to address the complex and evolving needs of a client and interact with said client on a subject that could range from as trivial as a parking ticket to as consequential as a murder charge? The short answer is one could not -- and therein lies, at least for the time being, one of the significant limitations of artificial intelligence as it relates to the law.
Assuming that one understands that a machine will probably not be taking the first meeting with a prospective client, what services can AI subsequently perform? AI is currently being used to review legal documents at a rate that is nothing short of astounding. According to Bloomberg (paywall), JPMorgan Chase announced that it had utilized COIN software for contract intelligence to review in seconds what had previously taken a slew of lawyers and loan officers some 360,000 hours to review.
If the magnitude of this computing power does not impress you, then consider the accuracy of the work performed by LawGeex, whose automated contract review platform answers the elementary question, "Can I sign this?" LawGeex has claimed that in February of this year, its artificial intelligence reviewed 5 NDAs to an accuracy of 94% in 26 seconds, compared to a team of 20 carbon-based lawyers who received an average accuracy score of 85% and on average completed the review of the same 5 NDAs in 92 minutes.
What does all of this mean to the practice of law and the future of the legal profession? Many would argue that a machine, irrespective of its power, could never replace the human attorney in terms of finesse, skill and intuition. Further, even assuming that a computer could be programmed to "think like a human lawyer," would we trust a computer to sit next to us and defend our lives, our livelihoods or our legacies in a court of law?
I think not only will the machine learn finesse, skill and intuition like a human, but sooner rather than later the machine will outthink human lawyers in traits that were once thought to be exclusive to our fellow carbon-based legal professionals.
In May 2017, Ki Jie -- widely considered the greatest Go player in the world -- lost three matches to the computer AlphaGo. What is perhaps most interesting about this event is that the nature of the game GO is such that there are more potential moves than atoms in the universe, and thus the computer was unable to gain any competitive advantage by employing brute force computations. How, then, did AlphaGo beat a human being in a game that relies heavily on intuition or the ability to understand something without the need for reasoning? Has the machine already surpassed our intuitive, emotional and instinctual thinking abilities?
Finally, would we trust a computer -- a robot -- with our lives on the line? The answer to this, I believe, is an unequivocal yes. In fact, we do it every single day. We fly on automated planes without giving it a second thought, and in the very near future, we will step into our automobiles and exit them without ever touching the steering wheel.
What, then, becomes of the human lawyer? Well, from this attorney's perspective, it leaves us at the dawn of the golden age of the profession. The business is changing -- the world is changing -- and AI is molding and shaping the future of almost every aspect of our lives, including the law. In the 16th century, Hans Geing depicted Lady Justice with a blindfold to represent her impartiality as it relates to wealth and power and now, almost 500 years later, that blindfold may well come in the form of a computer.
