ROBOLAWYERS? REALLY?
(Posted December 2, 2016) I assume you’re sitting as you read this. If you aren’t, pull up a comfortable chair, preferably one that won’t tip over easily.
I’ve just finished reading a fascinating new book, Megachange by Darrell M. West (Brookings Institution Press 2016). West analyzes recent historical trends and notes a pattern: we’re in an era of remarkably significant changes in many sectors of society. At just 171 pages (exclusive of endnotes), it’s an easy read, and I won’t spoil it by describing it in detail.
But one section of the final chapter grabbed me by the lapels and shook me. In that chapter, West peers into the trusty crystal ball to try to foresee future megatrends. In the section entitled, “Robots Take the Jobs,” he foresees the possibility of major unemployment due to automation. You may recall significant debate in the just-concluded presidential election over the question whether American manufacturing job losses are due to businesses’ relocations overseas, or to automation. West sees the latter as the greater contributor, and the greater long-term threat to full employment, and in turn, social stability.
I know what you’re thinking. Okay; but not lawyers. We exercise judgment in choosing among alternatives, and robots can’t do that. But West tells the story of a “challenge” by Amazon to see if designers could create a robot that’s able to “autonomously grab items from a shelf and place them in a tub.” Now the company uses 15,000 robots to do just that, making the humans who previously did the selecting involuntarily unemployed.
Yeah, yeah, you’re thinking. That’s low-level decisionmaking; nothing at all like what lawyers do. We have to be creative, exercise real judgment. No robot can do anything like that.
Well, actually …
Here’s a passage from p. 148 of West’s book that will get your attention:
Anthropologist Eitan Wilf of Hebrew University in Jerusalem … describes a “jazz-improvising humanoid robot marimba player” that can interpret music context and respond creatively to improvisations on the part of other performers. Designers can put it with a jazz band, and the robot will ad lib seamlessly without listeners being able to discern any difference with human performers.
Uh-oh. That’s getting painfully close. The Boss was a professional musician before she retired – an operatic contralto – and she has told me on occasion that opera singers and other classical musicians envy jazz performers because of their freedom to improvise. If a robot can pull off something as creative as jazz, are we in trouble?
There are examples in other fields. You’ll recall that in 1997, IBM’s computer Deep Blue took on reigning world chess champion Garry Kasparov in a match, and beat him. Kasparov was no ordinary player; according to the ratings I’ve seen, he had at the time the highest rating of any player in the history of the game. (Magnus Carlsen, who defended his world title in the past week, has since surpassed Kasparov’s late-90s rating.) Isn’t that creative thinking, at a profound level? The same company’s Watson took on the two most dominant players in the history of the game show Jeopardy!, and crushed them both.
An article in the Fall 2016 issue of Litigation, the ABA Lit Section’s quarterly magazine, which just arrived this week, reports that Thomson Reuters, the parent of Westlaw, is now using Watson in the legal industry. (Deep Blue is retired and sipping on WD-40 cocktails on a beach somewhere.) If you need a name for the cyber-interloper into our profession, Watson is as good as any.
But it’s still not what lawyers do. True. But go back in your mind’s eye 40 or 50 years and ask that generation’s auto workers if a robot could ever assemble a car. Go back even five years and ask those folks who thought they had secure jobs in the Amazon warehouse, if a robot could do what they did. A decade (or less) ago, could you find a taxi driver who would envision that a car could drive itself? Are you old enough to remember law practice in the 1970s, before Lexis and Westlaw changed the process of legal research? Who says our profession can’t change?
Better yet, pick your favorite search engine and – actually, go pour yourself a stiff bourbon and come back; go ahead, I’ll wait – search the phrase “robots replace lawyers.” I did it on Bing and got fifty million hits. I poked around some of the articles and found ominous titles, though the articles themselves usually included assurances that the robots would only replace lawyers for ordinary tasks like document reviewing; not for processes requiring deep thought. The Litigation article that I mentioned above concludes with this reassurance:
Attorneys concerned about being replaced by [Artificial Intelligence] need not fret because based on these thought leaders, AI is being developed not to replace attorneys but to help them.
Perhaps so; but some of the articles I surveyed were a tad more foreboding, as I’ll discuss in a moment. Some of the predictions included dramatic reductions in new-associate hires, since robots could handle time-intensive tasks without expecting vacations or consideration for partnership. (So if the market for new lawyers dries up, where’s the next generation of senior lawyers going to come from?)
But robots can’t develop the personal relationships that humans can. That’s true, too. And as long as humans are the clients, it will be an advantage to have a human in the law office. But some clients won’t care; if a robotic lawyer can undercut a human lawyer on price, some clients will give it a try, and if word spreads, who knows?
Even so, some experts believe that that human touch will become superfluous “within decades.” See this very recent article from the Harvard Business Review, forecasting that “the traditional professions will be dismantled, leaving most, but not all, professionals to be replaced by less-expert people, new types of experts, and high-performing systems.”
What does all this have to do with appellate lawyers? None of the articles I skimmed mentioned a robot’s giving an oral argument, and in truth, I can’t picture that, either. But as robotics technology continues to advance, who can say where the limit is? As I noted in a recent essay on oral argument, the great advocates of the Nineteenth and early Twentieth Centuries would find our current practice appalling, unthinkable, what with our drive-by oral arguments and expansive written briefs. Who’s to say that in the not-too-distant future, robotic lawyers won’t submit briefs to robotic judges to generate a calculated decision without the time-intensive tedium of oral arguments?
If you view this possibility as dystopian, I agree with you. Please note that I’m in no position to say that it’s inevitable. I’m just telling you that it’s not as impossible as it seems now.
What’s a human to do? Garry Kasparov and Ken Jennings have probably been asking that question of themselves for several years now, ever since they went up against machines and came away with silver medals.
In the short run, there’s probably nothing you need to do; if you’re in my generation and may be practicing for another 15-20 years, your practice is probably safe. The appellate courts that sit in Virginia will be manned by human jurists, not robots, over that time, and they’ll demand to hear a human lawyer argue each appeal.
Now, if you’re just starting a legal practice and you’re in your twenties, you may have a different career outlook by the 2060s. (As if the glut of lawyers and the current terrible job market weren’t enough of a challenge.) For you the best approach is probably to cultivate the qualities of “judgment, creativity, and empathy” – the ones identified in the Harvard Business Review article where humans cannot currently be replaced. As long as other humans are making the decision of which lawyer to hire, you’ll have an advantage that the current generation of robots can’t match.
Update December 6: I’m not the only lawyer who’s pondering this topic. See this notice for an upcoming webinar from the ABA, entitled, “Alternative Legal Providers – The Predator That Eats You Will Not Look Like You.”