By Paul Lippe and Daniel Martin Katz
In July, we talked about whether the change in law should be characterized as “Disruption, Eruption or Interruption?” This week, we drill down into one likely source of change, IBM’s Watson.
Lawyers have been thinking for a while about whether artificial intelligence would ever start to displace or complement lawyers. Richard Susskind, the leading legal futurist/technologist, did his work in this area starting in the mid-1980s. In the August issue of the ABA Journal, one of the commenters to an article about LegalZoom feared: “Once we have fully artificial intelligence enhanced programs like LegalZoom, there will be no need for lawyers, aside from the highly specialized and expensive large-law-firm variety.”
Similarly, August’s American Lawyer carried a long discussion about Watson. “One thing that is abundantly clear is that there are things that lawyers have traditionally done that can be automated,” Cooley partner Craig Jacoby told the American Lawyer. “There’s no sense fighting that.”
For a longer view of how these systems work, let’s quote the chief technology officer of Synopsys (Paul’s old company)—arguably was one of the more successful AI companies of the time—circa 1993: “It’s only AI when you don’t know how it works; once it works, it’s just software.”
As part of its thoughtful launch to deliver Watson’s cognitive computing capabilities through services, IBM has begun to partner with different companies in different fields, including law. (Full disclosure—Legal OnRamp is working with IBM on Watson with help from Dan.) This is an interesting go-to-market approach for IBM, to create an ecosystem around a nascent technology, with analogues to Google search or Apple’s iTunes (or for the true aficionados, to Salesforce’s Appforce).
Many imagine Watson might displace lawyers for legal reasoning. We believe that systems like Watson are very unlikely to displace the reasoning processes of lawyers. But it’s equally true Watson may illuminate how rare it is that lawyers have to solve “bespoke” reasoning problems, and how common it is to apply “proven” approaches in slightly different contexts. But Watson doesn’t have to displace legal reasoning to have an impact.
Here are 10 predictions:
• Watson is almost certainly the most significant technology ever to come to law, and it will give lawyers permission to think innovatively and open up the conversation about what is possible in a field that has been somewhat “stuck.” IBM and Sloan-Kettering have collaborated on a video talking about how Watson can help treat cancer patients better. We have no videos like that anywhere in law—maybe we should.
• Watson will force a much more rigorous conversation about the actual structure of legal knowledge. Statutes, regulations, how-to-guides, policies, contracts and of course case law don’t work together especially well, making it challenging for systems like Watson to interpret them. This Tower of Babel says as much about the complex way we create law as it does about the limitations of Watson.
• Watson will open up new possibilities (and challenges) for teaching. Within a few years, many or most of the Socratic method questions that get posed in a first-year contracts class will likely be answerable by students referencing their ContractsWatson at their desk. Many professors will hate this; most will recognize it as no different from the introduction of the calculator to algebra class in 1973.
• Watson will lead at least a few enlightened law schools to walk down the block to engineering schools to try to integrate other disciplines into the practice of law. (See “The MIT School of Law.”)
• Watson should make complexity more manageable concerning areas like Dodd-Frank, the Affordable Care Act, the Foreign Corrupt Practices Act, or mergers and acquisitions integration, reducing the cost of law.
• Watson will empower younger lawyers—who are traditionally at the bottom of the hierarchy and have now been dislocated by today’s job market—since they will likely be the first to embrace it.
• Watson will catalyze better organization of legal information and legal data, forcing organizations to better manage their current data and delivering substantial returns from this information management step alone. It will also help clarify what lawyers do and how they add value, and it will focus attention on the regulatory model for lawyers.
• Watson may be used as a dedicated or embedded service for specific legal workflows as much as it is as general purpose tool. Think how “smart” email programs now suggest possible addressees based on prior group emails.
• Watson (or something like it) will likely become a standard authoring/query model. Just as most companies today write their Web information to optimize for Google’s search, professional knowledge (which is published in a multi-tier structure) will want to be better synthesized through a system like Watson and will adopt new authoring and publishing norms.
• Watson won’t displace lawyers—it will make law more accessible and transparent, as it should be.
Watson, as well as other forms of machine learning, are likely to make further breakthrough in the direction of “quantitative legal prediction” in the years ahead. Forecasting is an important part of what many lawyers do. As an example, please consider a study recently released (by Dan and colleagues Josh Blackman and Mike Bommarito) that applies sophisticated techniques from machine learning to predict the voting behavior of the U.S. Supreme Court. Using only information available prior to the court’s decisions, their model was able to correctly predict 70 percent of the court’s ultimate outcomes.
While correctly predicting the Supreme Court is not a particularly commercially important activity, this effort points to how such techniques might be applied to other elements of law. Professors are teaching machine learning to law students, as they believe that going forward, expertise in analytics intersected with substantive legal knowledge is the source of significant value creation.
Twenty-five years ago, Synopsys brought AI to the field of electronic design. The consequence of that was not to put engineers out of business or replace them with computers, but to propel a virtuous circle of greater capabilities by improving humans’ ability to manage complexity.
Law is struggling with its ability to manage complexity. If Watson can help us do better, that’s a good thing.
Paul Lippe is the CEO of the Legal OnRamp, a Silicon Valley-based initiative founded in cooperation with Cisco Systems to improve legal quality and efficiency through collaboration, automation and process re-engineering.
Daniel Martin Katz, an associate professor at Michigan State University College of Law, is the director of the ReInvent Law Laboratory and the co-founder of LexPredict, a legal analytics consulting firm.