Lawyers and AI: Saying Susskind is Wrong

"Richard Susskind writes the same book every five years. He just updates the examples."

The above would be offensive if it were not a joke frequently delivered in the first person by Susskind himself. What makes the self-deprecation so humorous is that, in part, it is true. Richard Susskind has been delivering a fairly consistent message for decades. And he keeps finding more and more examples of the market making manifest his predictions about how technology will change the way legal services are delivered. Isn't that what being right looks like?

Prophesy can be a fairly ephemeral business. In 1996, Susskind was labeled "dangerous" and "possibly insane" because he was daft enough to posit that email would become the dominant form by which lawyers and clients would communicate. Within a decade, a lawyer was being laughed out of court after claiming "excusable neglect" for not checking his email regularly. These days, we write articles about lawyer overreliance on email. Apostasy became orthodoxy became barrier to innovation.

From email to the unbundling of legal services to the use of expert systems, Susskind has an excellent track record of outlandish predictions turning into common sense. I strongly recommend all his books, including his latest, The Future of the Professions. I tend to only disagree with Susskind's forecasts as a joke. Or, at least, so I thought until I read an interview with me where I went a bit further:
Flaherty – unlike, for example, Richard Susskind – is not pessimistic about the future for lawyers, quite the opposite. He believes that technology will not drive lawyers out of their role of trusted adviser. 'I think that technology can elevate lawyers' work to a higher level. I also think that younger lawyers will be relieved of the simple, brain-dead work. If you look at it this way, technology is a necessary precondition for allowing lawyers to be lawyers.'
There is much in the above that reflects what I think. But the interview--translated from Dutch--is not exactly the epitome of nuance.

First, I am somewhat neutral about the future for lawyers even if it is a future filled with lawyers. A lawyer-heavy world is not necessarily a good thing. I long ago stole from Susskind the idea that the law does not exist to keep lawyers in lucrative employment. And I am one of those lost souls who believes that our world has sacrificed too much talent to the unproductive sprawl of finance, tax and legal. I consider all those industries to be of substantial importance, but they are all susceptible to self-perpetuating arms races reliant on rent seeking and regulatory arbitrage. Whether this makes me a blinkered libertarian or unthinking socialist depends on your point of view.

Second, as I wrote here, I have no idea what is going to happen decades hence. I am not qualified to referee this point of contention between Susskind and others. I can't imagine a world much different from the status quo. I also can't imagine a scenario where the accelerating returns to technology do not fundamentally transform every aspect of work, including the work done by lawyers. And I can't imagine a world without work. Nor can I imagine what higher-value work it is that humans, including lawyers, will find to do if the automation paradox continues to hold and machine augmentation leads to a higher demand for human labor. Which is to say that I do not possess a fecund imagination. Humanity is fortunate that reality is not constrained by the limits of my imagination or intellect. Just because I have a hard time imagining something does not mean it won't/can't happen.

While I don't really have an opinion, or a rooting interest, on the net employment effects for legal professionals, I do think the nature of work will change. And, as expressed in the interview, I suspect it will change for better. That does not necessarily mean it will get better for everybody. As Tyler Cowen writes in Average is Over:
This imbalance in technological growth will have some surprising implications. For instance, workers more and more will come to be classified into two categories. The key questions will be: Are you good at working with intelligent machines or not? Are your skills a complement to the skills of the computer, or is the computer doing better without you? Worst of all, are you competing against the computer?
Third, while I find it intellectually fascinating, I don't derive much in the way of practical guidance from reaching firm conclusions about how the world will look a few decades from now. The Susskinds (Richard wrote the book with his son Daniel) predict the end of (most) lawyers, but not anytime soon:
Our expectation is that, over time— by which we mean decades, rather than overnight— there will be technological unemployment in the professions. In other words, there will not be sufficient growth in the types of professional task in which people, not machines, have the advantage to keep most professionals in full employment....

We cannot emphasize strongly enough that we are not predicting that the professions will disappear over the next few years. We are looking decades ahead in this chapter, and anticipating an incremental transformation and not an overnight revolution.
In the short and medium term, the Susskinds envision that we will be operating in a streamlined version of the present:
There are two possible futures for the professions. The first is reassuringly familiar. It is a more efficient version of what we already have today. On this model, professionals continue working much as they have done since the middle of the nineteenth century, but they heavily standardize and systematize their routine activities. They streamline their old ways of working. The second future is a very different proposition. It involves a transformation in the way that the expertise of professionals is made available in society. The introduction of a wide range of increasingly capable systems will, in various ways, displace much of the work of traditional professionals. In the short and medium terms, these two futures will be realized in parallel. In the long run, the second future will dominate, we will find new and better ways to share expertise in society, and our professions will steadily be dismantled. That is the conclusion to which this book leads.
From my reading, the long run is the point of disagreement. The counterprogramming tends to concur with Susskinds on the near-term implications of advancement in technology. Indeed, in Can Robots Be Lawyers?Professors Remus and Levy expressly limit their analysis to the next decade--a decade in which they, too, expect far-reaching effects from technological innovation:
Our focus is recent developments in legal automation, but we take as a given that earlier innovations dramatically impacted legal practice. Word processing revolutionized document drafting. The Internet permitted rapid document transmission and video conferencing; accelerated the breakdown of law firms’ information monopoly on rates, services, and clients; and increased clients’ ability to spread legal work among multiple law firms. Email increased the speed and ease of communication both among lawyers and between lawyers and clients, and expanded the number of associates a single partner could supervise and so has facilitated the growth of large law firms. These innovations changed law practice in fundamental ways. The next wave of technologies, our focus in this paper, promises similarly far-reaching effects.

We anchor our discussion in the current and foreseeable trajectory of these technologies in the present and near-term future (roughly the next decade). The resulting analysis is admittedly linear, risking that we underestimate the impact of radical future innovation.
Even if I were convinced of lawyerless future, I am not sure what the consequences of that conviction would be for me. Because of the aforementioned lack of imagination, I am hard pressed to conceive of a future of any indeterminate length that would affect my near-term thinking (an imminent apocalypse would change some priorities).

I, for example, would wager that we are headed for a future dominated primarily by driverless transportation. But there is a strong argument that I am wrong given the well-documented challenges of fully autonomous vehicles. Either way, I still need to drive and service my existing car, worry about road conditions, and remain concerned with whether my fellow drivers are paying attention. That will be the case until it isn't (if ever). My belief about the future--while I like to read and think about it--does not have much effect on my immediate present.

Similarly, if I were to be persuaded by Robert Gordon that progress has stagnated, I would advocate that the legal profession take advantage of the already available advances in process and technology to improve the delivery of legal services. If I were persuaded by Remus and Levy that technological advances are going to have far-reaching but not existential effects on the legal profession, I would advocate that the legal profession take advantage of the available advances in process and technology to improve the delivery of legal services. If I were persuaded by the Susskinds that the far-reaching impacts of technological advances over the next decade presage a future of dwindling lawyer employment, I would advocate that the legal profession take advantage of the available advances in process and technology to improve the delivery of legal services.

What came through in the interview was a bit of exasperation. My exasperation was not with the Susskinds or anybody else writing thoughtfully about, or working towards, introducing capable machines into the legal ecosystem. My exasperation was with the fact that I so often feel compelled to talk about robots replacing lawyers despite the fact that I don't even pretend to have any expertise on the topic. Moreover, I know that, in general, this discussion leads to an unproductive place, which is why so many of us caveat discussions of law and technology by assuaging fears that Skynet is on the verge of eradicating lawyers.

Too many people, a majority of whom are lawyers, treat the robots-replacing-lawyers question as if it is a binary condition. Either it will happen or it won't. Nothing else merits consideration. For them, there is no use in discussing the intervening decades where technology incrementally changes the way we work as we automate tasks rather than jobs.

On one level, this leads to a kind of existential dread and attendant hysteria where all people want is to be assured that their jobs are safe. If they don't get that assurance, they devote their mental energy to coming up with all the ways that humans are super special and can never be displaced. They latch onto clever turns of phrase like a commenter to my recent article, "a machine might know that a tomato is a fruit but a human would know not to put it in a fruit salad." You then freak them out with things like Chef Watson and plummet down the rabbit hole until you end up in "Can a submarine swim?" territory. But if they do get the assurance, they stop listening because the big question has been answered.

On another level, AI triumphalism--what someone much smarter than me calls "AI madness"--also stifles serious discussion. People buy into the hype. They start believing in magic and Easy Buttons. They just expect everything to work immediately and seamlessly. Intuitive interfaces (of which we have very few) stop being sufficient. They want machines that intuit the user's objectives. I want those machines, too. But they aren't quite here yet. For now, we still need to do the hard work of systems integration, security, training, workflow mapping, process design, and all the other things that people do not feel the need to discuss when they believe that technology will automagically solve every problem.

I'm all for measured discussions of the state of play of AI in law. I'm not dismissive of technological advancements. I concede that it is of genuine academic interest how those advancements will affect the profession a few decades from now. Truly, academics have a responsibility to think about the changing world for which they are preparing their students. But the rest of us have a responsibility to do better now. That means taking advantage of the imperfect technological advances that are already available to improve the way legal services are delivered. My hope is that we can do that while also enjoying and contemplating the provocative ideas that emanate from the Susskinds and their fellow travelers.

Casey Flaherty is the founder of Procertas. He is a lawyer, consultant, writer, and speaker focused on achieving the right business outcomes with the right people doing the right work the right way at the right price. Casey created the Service Delivery Review (f.k.a., the Legal Tech Audit), a strategic-sourcing tool that drives deeper supplier relationships by facilitating structured dialogue between law firms and clients. The SDR is premised on rigorous collaboration and the fact that law departments and law firms are not playing a zero sum game--i.e., there is more than enough slack in the legal market for clients to get higher quality work at lower cost while law firms increase profits via improved realizations.
The premise of the Service Delivery Review is that with people and pricing in place, process offers the real levers to drive continuous improvement. Proper collaboration means involving nontraditional stakeholders. A prime example is addressing the need for more training on existing technology. One obstacle is that traditional technology training methods are terribleCompetence-based assessments paired with synchronous, active learning offer a better path forward. Following these principles, Casey created the Legal Technology Assessment platform to reduce total training time, enhance training effectiveness, and deliver benchmarked results.
Connect with Casey on LinkedIn or follow him on Twitter (@DCaseyF).

Bookmark and Share


Jordan said...

Some great points here. While Susskind has said lots of things that have been true, it doesn't mean he's incapable of being wrong. We won't know what to expect from AI in law until the technology reaches a high enough level. Until then, all of us are just guessing.

Athelene Gosnell said...

I agree with your thoughts. Unfortunately, much thinking today is of the binary type you describe. That kind of thinking in the early 1900s would have people deciding that they should keep riding their horse until cars could fly. In the meantime, many of us have made pretty good use of the intermediate versions.


© 2014, All Rights Reserved.