In Machine Learning and Law, Harry Surden of the University of Colorado Law School provides a comprehensive and insightful account of the impact advances in artificial intelligence (AI) have had and likely will have on the practice of law. By AI, of course, Surden means the “soft” kind represented mostly through advancement in machine learning. The point is not that computers are employing human cognitive abilities, but rather that if they can employ algorithms and other computational power to reach answers and decisions like those humans make, and with equal or greater accuracy and speed, it doesn’t matter so much how they get there. Surden’s paper is highly recommended for its clear and cogent explanation of the forms and techniques of machine learning and how they could be applied in legal practice.
Surden quite reasonably recognizes that AI, at least as it stands today and in its likely trajectory for the foreseeable future, can only go so far in displacing the lawyer. As he puts it, “attorneys, for example, routinely combine abstract reasoning and problem solving skills in environments of legal and factual uncertainty.” The thrust of Surden’s paper, therefore, is how AI can facilitate lawyers in exercising those abilities, such as by finding patterns in complex factual and legal data sets that would be difficult for a human to detect, or in enhancing predictive capacity for risk management and litigation outcome assessments.
What Surden is getting at, in short, is that there seems to be little chance in the near future that AI can replicate the “bespoke lawyer.” That term is used throughout the commentary on the “new normal” in legal practice (which is actually a “post normal” given we have not reached any sort of equilibrium). But it is not usually unpacked any further than that, as if we all know intuitively what bespoke lawyering is.
To take a different perspective on bespoke lawyering and the impact of AI, I suggest we turn Surden’s approach around by outlining what is bespoke about bespoke lawyering and then think about how AI can help. In the broadest sense, bespoke lawyering involves a skill set that draws heavily from diverse and deep experience, astute observation, sound judgment, and the ability to make decisions. Some of that can be learned in life, but some is part of a person’s more complex fabric—you either have it or you don’t. If you do have these qualities under your command, however, you have a good shot at attaining that bespoke lawyer status. Here’s a stab at breaking down what such a lawyer does well:
Outcome Prediction: Prediction of litigation, transaction, and compliance outcomes is, of course, what clients want dearly from their lawyers. On this front AI seems to have made the most progress, with outfits like Lex Machina and LexisNexis’s Verdict & Settlement Analyzer building enormous databases of litigation histories and applying advanced analytics to tease out how a postulated scenario might fare.
Analogical and evaluative legal search: Once that pile of search results comes back from Lexis or Westlaw (or Ravel Law or Case Text), the lawyer’s job is to sort through and find those that best fit the need. Much as it is used in e-discovery, AI could employed to facilitate that process through machine learning. This might not be cost-effective, as often the selection of cases and other materials must be completed quickly and from relatively small sets of results. Also, the strength of fit is often a qualitative judgment, and identifying useful analogies, say between a securities case and an environmental law case, is a nuanced cognitive ability. Nevertheless, if a lawyer were to “train” algorithms over time as he or she engages in years of research in a field, and if all the lawyers in the practice group did the same, AI could very well become a personalized advanced research tool making the research process substantially more efficient and effective.
Risk management: Whereas outcome prediction is usually a one-off call, managing litigation, transaction, and compliance outcomes over time requires a sense of how to identify manage risk. Kiiac’s foray into document benchmarking is an example of how AI might enhance risk management, allowing evaluation of massive transactional regime histories for, say, commercial real estate developers, to detect loss or litigation risk patterns under different contractual terms.
Strategic planning: Lawyers engage extensively in strategic planning for clients. Where to file suit? How hard to negotiate a contract term? Should we to disclose compliance information? Naturally, it would be nice to know how different alternatives have fared in similar situations. Here again, AI could be employed to detect those patterns from massive databases of transactions, litigation, and compliance scenarios.
Judgment (and judging): Judgment about what a client should do, or about how to decide a case when judge, involve senses not easily captured by AI, such as fairness, honesty, equity, and justice. The unique facts of a case may call for departure from the pattern of outcomes based on one of these sensibilities. Yet doctrines do exist to capture some of these qualities, such as equitable estoppel, apportionment of liability, and even departure from sentencing guidelines, and these doctrines exhibit patterns in outcomes that may be useful for lawyers and judges to grasp in granular detail. What is equitable or just, in other words, is not an entirely ad hoc decision. AI could be used to decipher such patterns and suggest how off the mark a judgment under consideration would be.
Legal reform: As I tell my 1L Property students, in almost every case we cover some lawyer was arguing for legal reform—a change in doctrine, a change in statutory interpretation, striking down an agency rule, and so on. And of course legislatures and agencies, when they are functional, are often in the business of changing the law. To some extent arguments for reform go against the grain of existing patterns, although in some cases they pick up on an emerging trend. They also rely heavily on policy bases for law, such as equity, efficiency, and legitimacy. In all cases, though the argument has to be that there is something “broken” about continuing to apply the existing law, or to not invent new law, in the particular case or broader issue in play. AI might be particularly useful as a way of building that argument, such as by demonstrating a pattern of inefficient results from existing doctrine, or detecting strong social objection to an existing law.
Trendspotting: In my view the very best lawyers—the most bespoke—are those ahead of the game—the trendspotters. What is the next wave of litigation? Where is the agency headed with regulation? Which law or doctrine is beginning to get out of synch with social reality? Spotting these trends requires the lawyer to get his or her head outside the law. Here, I think, AI might be most effective in assisting the bespoke lawyer. A plaintiffs firm, for example, might use AI to monitor social media to identify trends highly associated with the advent of new litigation claims, such as people complaining on Twitter about a product. Similarly, this approach could be used to inform any of the lawyer functions outlined above.
Handling people: Ultimately, a top lawyer builds personal relationships with colleagues, peers, and clients. AI can’t help you do that, I don’t think, but by helping lawyers do all of the above it may free up time for a game of golf (tennis for me) with a client!
Many, many years ago, when I was practicing environmental law with Fulbright & Jaworski in Austin, I was unfortunate enough to have a number of clients whose needs required that I master the EPA’s utterly convoluted definition of solid and hazardous waste. One summer I assigned a summer associate the task of flowcharting the definition. Over the course of the summer we debugged draft after draft until, finally, we had a handwritten flowchart that flawlessly worked any scenario through the definition step-by-step. It was ten legal-sized, taped-together pages long. It worked, but it wasn’t very practical.
If only we had had Neota Logic back then! Last week, in my Law 2050 class, Kevin Mulcahy, Director of Education for Neota, demoed their product over the course of two classes and a 3-hour evening workshop. Prior to the session I had assigned the class the exercise of flowcharting the copyright law of academic fair use. Each student prepared a flowchart and explained its logic, then six groups collaborated on final work products. I sent the group flowcharts to Kevin so he could use them to explain the Neota platform in a context familiar to the students.
Neota is a software program that allows the user to translate legal (or other) content into a user-friendly interactive application environment, much like Turbo Tax does for tax preparation. Neota allows the content expert to build the app with no coding expertise, with end products that are quite sophisticated in terms of what can be embedded in the app and how smoothly the app walks the user through the compliance logic. Example apps Kevin offered covered topics as varied as songwriter rights to Dodd-Frank compliance.
The first class period Kevin introduced Neota and then walked through each of the group flowcharts to analyze how each one broke down the fair use compliance problem. The core theme was how important it is to develop the output scenarios first. In the fair use exercise, there are several yes/no questions specific to educational uses, and then a multi-factored balancing test applies in the event none of those binary questions leads to a fair use outcome. Like any balancing test, this one yields a range of scenarios from very likely fair use to very likely not fair use. We spent a good deal of time thinking about how to design an app component to capture the balancing test.
In the evening workshop a group of 20 students acted as content experts to guide Kevin through the process of building the fair use app, much in the way a legal expert might work worth a Neota software expert. The most striking learning experience from this session, besides the deep look under Neota’s hood, was how the process of building the app actually sharpened our fair use compliance logic. We tested various approaches for capturing the balancing test and conveying output scenarios with substantive explanations for the user.
The next day the entire class regrouped to go over the workshop product, allowing those who could not make the workshop due to conflicting classes the chance to get a good feel for both the flexibility and precision the Neota software offers. Thinking back to my perfectly accurate but impractical ten-page flowchart of the EPA’s waste definition, I could envision how that and many other tasks that required developing a compliance logic could have been leveraged into apps I could have shared with other attorneys in my firm as well as clients.
My Law 2050 students clearly got a lot out of the immersion in using Neota to attack a compliance logic problem. I can’t thank Kevin and Neota enough for the time he invested in preparing for and delivering what was an excellent hands-on and instructive workshop. By the way, the EPA now has an online decision tool for navigating through the waste definition. I think they might want to get in touch with Neota!
This week my Law 2050 class has been all about Lex Machina, and to quote one student at the end of the two sessions: “I can’t imagine being a patent law firm and not wanting to purchase that!” [Note: I have no connection whatsoever with Lex Machina other than having them appear in my class, nor, I believe, did this student.] That sentiment was widely shared.
I contacted Lex Machina early in the semester to explore how I could give the class a deep dive in their technology. Jeremy Mulder, Lex Machina’s Director of Customer Success, worked closely with me to make the site available to the students, design an exercise for us to complete in one class, and guide us through the site and the company’s vision over a JoinMe link the next day.
First, the Lex Machina product is a truly awesome example of turning Big Data into a useful, user-friendly legal analytics product. The depth and breadth of data contained in the site, particularly for patent law, was astounding. For example, pick any federal district judge and within a few seconds the site provides an array of data, including outcomes at granular levels, patents handled, time to case termination, lawyers appearing in the court, and many more. The site display and navigation is a breeze. The class started to tackle the questions together at the beginning of the first class, and within about 10 minutes, with no instructions from Lex Machina, we had begun to navigate the site with ease and, over time, learned how to tap into one after the other of analytic tools. The site is a model for other law+tech developers.
Second, as the exercise progressed I began to wonder how I would describe Lex Machina within the “disruptive technology” space. Disruption comes in many forms, and whether good or bad depends on the beholder. Lex Machina strikes me as disruptive primarily by providing an additive function—it makes possible what a lawyer could not have imagined he or she could do, at least without a tremendous amount of effort, time, and cost. It adds a tool, but it does not necessarily replace lawyers, or suck away billable hours, or “commoditize” a lawyering function; indeed, by giving lawyers more power over how to analyze patent law’s expanse, it may do just the opposite. More on the “disaggregation” of the disruptive legal technology concept into more descriptive and refined categories in an upcoming post.
The October 2014 issue of The American Lawyer includes its annual rundown of the Global 100 – the top 100 law firms around the world based on revenue – and I must say there’s quite a bit of legal spend going on out there! My ballpark estimate of the total revenue of the top 100 firms is close to $100 billion. Profits per partner at the Global 100 averaged $1.61 million in 2013, which is up (yes, up) 5.3 percent over 2012 (which was up 0.7 percent over 2011), with 70 firms averaging over $1 million. As expected, US firms took the biggest share of profits. Interestingly, some of the largest firms in the world in terms of revenue and/or lawyers fall below the $1 million PPP mark, including Jones Day, K&L Gates, Norton Rose Fulbright, Squire Sanders, CMS Legal, and Dentons. In some cases that may be a tradeoff between stability and profit, and in some cases the verein structure could result in uneven PPP across the firm’s offices.
As one might expect, the lion’s share of all that legal spend goes to firms with their largest number of attorneys based in the US. Of the top 50 firms in the world, 42 find their largest number of attorneys in the US. For the top 100 firms the number is 78, with the UK a distant second. In some cases, however, the US share, while the largest for the firm, is still relatively low. Overall, though, over 97,000 of the lawyers working for the Global 100 work in the US. The UK comes in a distant second with 6900 and China is third with just over 2500. But the number of lawyers working for AmLaw 200 firms in offices outside of the US, and the number of offices outside the US, has expanded steadily since 1998, primarily into the UK, but also significantly into China, Germany, France, Australia, and Canada.
The bottom line: Global Big Law seems to be doing just fine and spreading its wings. Is news of its death premature?
Head to this Computational Legal Studies site post for an excellent Prezi by Nicola Lettieri of the University of Sannio on legal and other issues flowing from the “Big Data Deluge.” One part of the presentation dives deep into the field of law and computational social science, which IMHO is the future of legal research and scholarship.
The second Network Analysis in Law conference will take place in Krakow, Poland, December 10-12. The call for papers outlines intriguing lines of research about legal networks:
We invite papers and demonstrations of original works on the following aspects of network analysis in the legal field:
- Analysis and visualization of networks of people and institutions: law is made by people, about and for people and institutions. These people or institutions form networks, be it academic scholars, criminals or public bodies and these networks can be detected, mapped, analysed and visualised. Can we better study institutions and their activities by analysing their internal structure or the network of their relations? Does it help in finding the ‘capo di tutti i capi’ in organized crime?
- Analysis and visualization of the network of law: law itself forms networks. Sources of law refer to other sources of law and together constitute (part of) the core of the legal system. In the same way as above, we can represent, analyse and visualise this network. Can it help in determining the authority of case law or the likelihood a decision will be overruled? Does it shed light on complex or problematic parts of legislation? Is it possible to exploit networks visualization to support legal analysis and information retrieval?
- Combination of the first and second aspects: people or institutions create sources of law or appear in them: Research on the network of one may shed light on the other. Two examples: (1) Legal scholars write commentaries on proposed legislation or court decisions. Sometimes they write these together. These commentaries may provide information on the network of scholars; the position of an author in the network of scholars may provide information on the authority of the comment. (2) The application of network analysis techniques to court decisions and proceedings is proving to be helpful in detecting criminal organizations and in analysing their structure and evolution over time.
I wish I could go!
Internet Millionaires: How Crowdfunding’s Viral Popularity Foreshadows a Future Need for Regulatory Compliance
Guest Post by Law 2050 Student Alex Nunn
As social media membership rates continually push to record heights, an emerging new trend is now seeking to turn your friends list into a pool of potential investors. “Crowdfunding,” as the movement has been coined, is the practice of raising capital by appealing directly to a large group of potential investors via the Internet. While the viability of such a trend might initially be met with skepticism, the equity-raising potential of crowdfunding has proven substantial. For example, in October 2012, Cloud Imperial Games pitched Star Citizen, a space combat video game, to the public and sought to raise the necessary capital for the game’s production through online crowdfunding. The idea quickly went viral, and by August 2014, the developer had raised $52 million dollars for its project, with over five hundred thousands individuals chipping in. On the more comical side, one individual used a crowdfunding site to raise $55,492 to help in his quest to make himself potato salad, while another start-up has raised £8,016 towards its mission to manufacture and sell giant inflatable sculptures of Lionel Richie’s head.
Undeniably, crowdfunding is attracting a significant amount of attention from prospective investors and commentators alike. Recently, however, the trend has caught the eye of a much more influential force – the United States government. Over opposition from the Securities and Exchange Commission, Congress passed the Jump-start Our Business Start-ups Act, or JOBS Act in 2012, which mandated regulatory support for crowdfunding. While the ability to quickly raise capital spurs on the current administration’s drive to bolster small business, the SEC remains wary of the movement due to the certain dangers that accompany crowdfunding.
For one, venture capitalism (the more formal method through which new businesses raise start-up funds) is an extremely risky endeavor for financial experts, with over eighty percent of start-ups failing in their first year. If even these seasoned financial professionals struggle to effectively predict the potential success of future start-ups, how much more vulnerable might crowdfunders be? Despite their enthusiasm, there exists a great potential for loss.
More importantly, crowdfunding is ripe for fraud. Through their crowdfunding campaigns, individuals can raise substantial sums without providing any identification, disclosure, or transparency with their plans. For example, despite its seemingly obvious unrealistic nature, an individual raised over $18,000 to manufacture his proposed “home quantum energy generator.” Predictably, his initial promise of free energy has yet to be fulfilled.
As crowdfunding grows out of its infancy, the movement’s stakeholders will increasingly demand legal aid. As one commentator notes, potential issues include questions over whether a crowdfunded start-up will be required to provide audited financial statements, and whether the funders, or even the funding portal, may share in any potential liability caused by the prospective campaign. Ultimately, the soaring popularity of crowdfunding will see a significant increase in the demand for regulatory compliance, especially as the SEC works towards issuing its final crowdfunding rules. As unassuming individuals find themselves on the receiving end of millions of dollars, their very first need, even before they begin to construct inflatable Lionel Richie sculptures, will be for sound advice on how to manage their funds in a safe, legal manner.
Guest Post by 2050 student Catherine Moreton
Tech company ComSonics announced in September that it is developing a new type of radar gun that detects not speeding, but texting. ComSonics specializes in handheld radar devices used mostly by cable companies searching for emission leaks in broken wires. But at the second annual Virginia Distracted Driving Summit, ComSonics revealed that the same technology is being adapted to track radio frequencies emitted when a driver sends a text message. According to spokesperson Malcolm McIntyre, the device can distinguish between frequencies emitted by text messages and those emitted by phone calls or emails.
In a year that included the National Highway Traffic Safety Administration (“NHTSA”) launching its first-ever national advertising campaign against distracted driving and AT&T’s “It can wait” campaign going viral, overdue public awareness of the dangers of texting and driving has increased dramatically. This is wonderful news for road safety and the 44 states (plus D.C.) that have banned texting while driving. But at what cost should we allow police officers to enforce those statutes more directly?
While McIntyre says the radar gun is “close to production,” technological concerns range from how to pinpoint whether the driver or a passenger was the one texting to what to do about automatic response messages. The technology is also currently limited to SMS messages and cannot yet detect texts sent over Wi-fi between iOS devices. Absent a safe harbor, the government might eliminate this boost for smartphone owners using the Communications Assistance for Law Enforcement Act (“CALEA”) to require providers to enable detection.
Once those kinks are worked out, privacy law will take center stage. McIntyre insists that the technology cannot decrypt the content of the messages, and under conventional Smith v. Maryland wisdom, this distinction would limit the government’s Fourth Amendment liability. But in a 2012 concurrence, Justice Sotomayor began to poke holes in the applicability of Smith to cell phone cases, calling the third-party doctrine “ill suited to the digital age.”
Plus, even though a cell phone user never reasonably expects her metadata to be private, the best evidence that a driver was texting is the time-stamped text itself. And if police want a driver to hand over her phone and incidentally reveal its contents, two June 2014 Supreme Court rulings suggest they’re going to need a warrant.
This summer, Riley v. California and companion case United States v. Wurie made huge advances for individual data privacy rights regarding cell phones, requiring a warrant for police to search essentially any kind of cell phone. With those opinions, the Supreme Court granted digital devices full Fourth Amendment protection absent exigent circumstances.
It is yet to be seen how Riley will affect other privacy arguments that could challenge the radar guns. Kyllo v. United States (while distinguishable in that it had to do with a home, which carries the strongest expectation of privacy) could require a warrant until the radar guns are “in general public use.” Independently of Fourth Amendment causes of action, the Stored Communications Act (“SCA”) could provide a remedy for phones searched without a warrant, and the Pen Register Act could require a court order before the radar guns may be used at all.
The good news is that the Court, after some resistance, seems ready to embrace the challenges of the digital age by beginning to agree with Justice Scalia’s 2010 view that “applying the Fourth Amendment to new technologies may sometimes be difficult, but when it is necessary to decide a case we have no choice.”
Given how much time we spend in law school covering what the law was and is, one of the goals of my Law 2050 class is to get students to think about what the law will be and how they can help shape it’s future. I have students identify examples of two kinds of trends. The first is an “inside law” trend, such as new technology and new kinds of service providers, that will influence how law is practiced. The other is an “outside law” trend, such as developments in health care, technology, and the economy, that will influence how law evolves in response.
Last year I had students work in groups to present “pitches” in a shark-tank setting, with the pitch being an assessment of whether to invest in the trend (e.g., put money into a new legal practice technology or devote firm resources to developing a new practice area). This year I have used this phase of the class to develop some practical, practice-oriented writing skills: a blog post, a client alert letter, and a bar journal article. As was the case last year, once again I am thoroughly impressed with the topics the students selected, and their blog post assignments were top-notch. Watch for several of them in coming days as students serve as contributing bloggers!
Here’s a sample of the topics:
Inside Law Trends: lawyer coaching for pro se clients; IP prior art search outsourcing; third party litigation funding; Shake, the contract app; legal hackathons; legal fee analytics; Ravel Law; Mitratech’s software for in-house counsel; “low bono” law firms; legal project management firms; online dispute resolution; pricing consultants; Islamic finance practice; speech recognition programs for lawyers; Bryan Cave’s Rosetta project; legal knowledge engineering; telecommuting and the decline of the law office; Counsel on Call; Integron; business for lawyers training programs; legal solution engineers; Clerky; Axiom–is it becoming another BigLaw?; virtual courts; Legal Force; and compliance lawyering.
Outside Law Trends: digital signatures; commercial delivery drones; invisibility cloaking; Google Glass; neural implants; predictive policing; driverless cars; commercial space travel; e-money; The Internet of Things (embedded sensor networks); newsgathering drones; unmanned cargo ships; virtual patient consultations; 3D printing of guns and organs; apps to convert 3D iPhone photos to 3D printing; Apple’s fitness watch; automobile connectivity and privacy issues; texting detection technology for police; cloud storage issues; sea level rise; crowdfunding; negligent infliction of disease; ridesharing (Uber etc.); robotic surgery; renewable energy trends; extreme reality TV; fracking; human gene patenting; and police body cameras.
Needless to say, we are going to have some interesting class discussions!