The Juris Blog.

Juris is making the world more fair through law-powered technology. Here you'll find project updates, research, and discussion about legal tech and access to justice.

Droids Not Robots – How most legal market automation numbers miss the point.

As great a headline as it would be, this isn’t about “robot lawyers” coming to take jobs. It’s about R2D2 following you around chirping pleasantly.

Article Author:

Adam Kerpelman

Adam Kerpelman

Co-Founder, President, and CEO at Juris

Like all the best posts, this also starts with a chat on Twitter:

Scary Headlines

There, I was tweeting about a CNBC piece talking about how technology is changing legal work, and how law schools are responding. The crux of the piece is a statistic from a McKinsey report predicting that 23% of “legal” occupations could be automated. The rest is about how law schools are shifting what they teach in order to compensate. 

The point of the CNBC piece is right on. Automation is coming, and professional schools are adapting to the changing professional landscape, as they always have. But, they’re missing the point on that 23%. Which is too bad for them, because 23% feels way too low. (They could have had a much higher number in that headline. Yay clicks!)

What they’re missing starts a step further back from the number a the question, “why do we care?” No one cares that automation replaced my hand saw. We care because these are jobs numbers, and that is important to acknowledge. Viscerally, these are numbers about how people feed their families. But broadly they matter because people want to understand the business landscape so they know where to look (or train) for jobs. Because of this, scary headlines work really well, as does my saying that number should be higher. 

Well, I’m happy to say that the lawyers reading don’t have to worry, because that “23% of lawyer jobs getting automated” idea doesn’t really say much about the business landscape that the schools are adapting to. That landscape is about law firms, and ultimately about the people they help, not just “lawyers.” And while that automation number should be way higher if they meant to portray the whole picture, the volume of work is only going to increase because of this type of automation.

The Business Landscape

It’s clear from the McKinsey study that when they say “legal occupation” they only mean lawyers, so right off the bat this isn’t any sort of representation of the actual legal market. This is literally just people who get paid to be a lawyer, and 23% of what they do can (arguably) be replaced. Anyone who has visited a law firm, even a fairly small one, can see that there are many more people in that building than just the lawyers. That same McKinsey study has office and admin staff sitting somewhere around 85% automate-able. How much of the firm is that

When we look at the firm structure, and the gap between that 85% and 23% we realize that the whole mental model here is kind of broken. It’s a fair model. Sometimes technology works like this. It comes and takes jobs from people, and in many fields those people have nowhere else to go. But, more often, technology creates more jobs than it takes. And in a legal market where ~200 Million US Citizens aren’t getting the legal help they need, we at Juris believe that legal will play out like this: technology will take some jobs, but it will grow the pie so much that even at 85% automation the legal services sector will create more jobs and earn more money than it ever has.

But this isn’t about The Terminator, it’s about R2D2. Because of the nature of the law the human element simply isn’t going anywhere. Which is why we get to talk about centaurs before we get to droids!

This game has centaurs. (How D&D is that?)

Law is complicated, which means what lawyers do all day is complicated. Within the law there are rules, but there are also a lot of moving pieces. You know what else is complicated? Games like Go and Chess. (Sorry Starcraft, don’t have time to explain that one to them now.)

Let’s start with Go. It’s a strategic game that is 1 million trillion trillion trillion trillion times more complicated than Chess. We can run those numbers because the game boards are finite. We know how many “possible moves” there are. This many moves makes Go more “strategic”, whereas Chess is more “tactical”. Across games, as the number of possible moves increases the balance between strategy (what you should do) and tactics (what you can do) shifts toward strategy. But they’re both there. Always. Where there are rules there will be tactics. Computers are really good at tactics.

Within the last few years computers have surpassed the human ability to play Go. A couple decades ago a computer beat Garry Kasparov, the world Chess champion. We haven’t beaten them at Chess since. Not sure where we are on Go, but if you followed the link above, you’d know that the world champion retired, proclaiming AlphaGo unbeatable. 🤷‍♂️

(Not in a rush? watch this:)

Humans still play chess, But, they play what’s called “hybrid” chess. (Or sometimes called Centaur Chess, which is way cooler.) They play with a computer by their side offering tactical suggestions all along the way, and the human makes strategic decisions as to the application of those tactics. You’d think that this would collapse into boring and repetitive games, but it doesn’t. Even chess is complex enough that as long as there are humans, there’s strategic variation.

So, lets ramp up to max complexity. Real life conflicts.

The gameboard for the legal professional is human life. We have rules. We’ve written them down. We call them laws. So, we have a gameboard, with rules, which generates a number of possible moves in any given scenario. Lawyers call those “fact patterns” and we study the shit out of them in school to understand how to decode them and respond.

Because there are so many moves, it’s easy to say “this isn’t Chess.” But, it might be Go. There really are only a finite number of moves under the law a lot of the time. Either way, it does boil down to the same dynamics: tactics and strategy. Games like Chess and Go teach us these things, and in turn we’ve used them to teach computers. And computers have gotten really good. Call it an iPhone, an app, a robot, or a droid, it sure seems like this ends with a computer sitting next to us.

If the future is hybrid lawyers, what does the legal market look like? 

The New Legal Market, and Pie for Everyone

Let’s get to the real questions you’re here for: who is going to lose jobs? Who is going to need to shift skillset? How do we teach the next generation? Let’s look back to the numbers from the study, it’s only 23% of lawyers, but it’s 85% of “office and admin” staff. This means somewhere in-between is a look at what actually happens to the average law firm.

At Juris, we don’t think this plays out like robots replacing jobs, we think it plays out as droids for everyone. Droids for normal people for the easy tactical stuff. Droids for professionals for the increasingly complex strategy that comes with escalation. And there are many cool projects working on various prongs of this problem. But, let’s talk about how this isn’t “coming for your jobs.”

Thus far, we’ve really only talked about what this looks like for lawyers, with their little droids following them around in the courtroom. But the reality is that the job of lawyers is to help people with their disputes. This is the real product of the complexity of real life. Normal people need other people to help them, eventually. There is no version of this system without humans helping humans, we just don’t need them all the time. It’s stupid to pretend we do, which is sort of the current charade, and it’s even more stupid to fight as if there isn’t enough work to go around.

In a business where 78% of demand is left unmet in the United States (nevermind global volume) we see a legal system that has not been able to scale. As such, it is manifestly unfair, and unequal. At Juris, our obsession is leveling the legal playing field for the average person, in order to unlock this scale. And the key to unlocking this scale isn’t droids for lawyers, it’s droids for people, and they won’t replace lawyers or law firms.

The people we aim to help with Juris weren’t going to get a lawyer anyway. Your average person is not bringing in a lawyer to help get their security deposit back. By building tools to allow people to defend their rights we are drawing new people into the legal system. They’re not using a lawyer (yet,) but they were never going to anyway. This case wasn’t complex enough. So, no jobs stolen here. But, there will be escalation of some of these cases, because that’s how the rules of the legal system work. Some percent of these cases will move to needing help from other humans. So, there it is, tech just created more work for lawyers. Out of thin air, I guess? 

The main point: see how the pie grows for everyone if we focus on people first and build tools that scale? This isn’t the industrial revolution version of robots coming for our jobs. This is the version where scale unlocks more jobs than ever.

Want to see a legal service droid in action already? Use our DepositLetter to get your security deposit back and think of it as a cheerful little blue droid chirping at you pleasantly to help you with your legal problems.

Juris is not a law firm or a substitute for the advice of an attorney. Juris provides a self-help service with general information and assistance only. Juris cannot and does not provide legal advice. Your access to this website and use of this service is subject to our Terms of Service.

If you have any comments, questions, or just want to say hi 👋, we would love to hear from you! You can use the chat system in the bottom right corner of this screen or email the team directly at