In a previous article I explored how Ethan Mollick’s insights from Co-Intelligence: Living and Working With AI about how the equalization of skills will impact legal. In today’s post I will examine two additional Mollick insights that are equally important to the profession.
The Role of Experts
Mollick notes in his book that AI tools have the capacity to create enhanced expertise on anything and everything by everyone and anyone. This dynamic is already playing out in legal practice. Want to be an expert on non disclosure agreements? You can do several AI prompts and get much of what you need to know to get by, at least for more routine questions.
Mollick insists, however, that expertise in most fields will still be necessary and critical. If nothing else, there needs to be experts to determine the accuracy and completeness of AI outputs, to catch the hallucinations, to see when an output is simply off mark or not relevant to the situation. To apply the information effectively.
Where will these experts come from when AI already has access to all the expertise that exists? How can we insure a pipeline for future experts? Mollick believes that in the near future, experts will be developed in new and different ways. AI itself will play a role in developing the new experts of the world.
And this expert development challenge is particularly acute in legal practice, where traditional training methods are already being questioned.
The Role of AI In Training
Historically, the guild system was used to train lawyers and help them become experts. The system works like this: the young lawyer graduates from law school and goes to work for an experienced lawyer. The young lawyer potentially sits at the feet of the experienced lawyer and learns through observation and undertaking assigned tasks under the experienced lawyer’s supervision.
By seeing and experiencing various scenarios and undertaking tasks repeatedly, the young lawyer learns to engage in critical thinking, assess strategy, envision the outcomes from various actions, and read clients and situations. The young lawyer thus gradually obtains the wisdom—gut instinct—of the older, more experienced lawyer.
That system is time honored but time consuming. And it is often inconsistent in its results. Some young lawyers work for good mentors and teachers and advance. Other young lawyers who might have the same raw ability do not and flounder.
AI disrupts that system. Many of the tasks that young lawyers historically undertook in their development cycle will, in the future, be done by AI. As I have written before, they will no longer have the in person opportunity to learn through conversation and constructive criticism.
But if Mollick is correct, there may be an answer. AI itself could ultimately become the mentor, the experienced voice in the room. It could be the aid and assistant to help the young lawyer develop the very skills that the traditional system promoted. This would involve using AI tools in the right way: not merely telling the young lawyer to use AI to get the answer, but giving the young lawyer a problem and asking the lawyer to create a solution. Only then would AI be used for suggestions to improve the solution. This is much like what an experienced lawyer does now.
Some Examples
For example, if assigned to prepare a list of questions for a deposition of a corporate representative, one option would be for the young lawyer to ask a large language model for the questions, which would produce a list from which the young lawyer could select. However, this does not necessarily train the young lawyer to identify the best questions for any given situation.
A more effective approach would be for the young lawyer to first develop the list of questions and then ask the AI tool for a critique, provided of course that the AI tool could do that. And this way, young lawyers could differentiate themselves in an AI-driven world by having critical thinking skills and the ability to apply those skills to human relationships and differences. As Mollick points out, the AI tools are not yet at this point. However, he believes they soon will be.
The urgency of addressing these training challenges becomes clear when we consider Mollick’s final insight.
AI Is the Worst It Will Ever Be
Mollick encourages us to assume that the AI we have today will be the worst AI we ever have. In other words, AI will develop perhaps exponentially, but certainly over time. Mollick says we cannot envision AI and what it will be tomorrow. Indeed just think how far we have come since the GenAI went public in November 2022 and how far its come in the last year.
This concept too has implications for legal. We should begin thinking now about the possibility of AI becoming a valued coach in the development of young lawyers and their skills. We should be planning and thinking about how that training looks, what it needs to involve, and what the results and measurables will be. Otherwise, we will be left with lawyers who know little about the practice of law other than to prompt AI for a list of deposition questions, which may or may not be relevant given the facts of any case.
What Does this Mean?
Lawyers and law firms need to be aware of the enormous implications of AI and its impact. These last two insights along with the idea that AI will equalize skills suggest that the legal profession is at an inflection point. The lawyers who thrive will be those who understand that AI levels the playing field, that expertise still matters but its development looks different, and that the tools we have today are only just the beginning.
The question for every lawyer and law firm is: Are you preparing for this future, or still hoping it won’t arrive?
Stephen Embry is a lawyer, speaker, blogger, and writer. He publishes TechLaw Crossroads, a blog devoted to the examination of the tension between technology, the law, and the practice of law.
The post Insights On AI And Its Impact On Legal, Part Two appeared first on Above the Law.