Podcast Ep #67: AI & Legal Writing: How to Use It Effectively with Jonah Perlin

We all know that one of the core skills of lawyering isn't having all the answers, it's knowing how to get to them. And probably the best way to do that is through writing. In today's episode, I welcome Professor Jonah Perlin to the program. He teaches in Georgetown Law School's legal practice program where he lays the foundation for first-year law students to become effective legal communicators. But his lessons are essential reminders for lawyers at every stage of your career.

Jonah shares the two key things that are essential to effective communication, written or otherwise. And we also explore how putting ideas into words and wrestling with language is the best path to stronger analysis and better understanding. And that learning process could be at risk if you turn too much of your writing over to technology like generative AI. But Jonah also shares how he's adopted AI in his own writing. He's got a cool trick for accelerating feedback loops and getting a specific type of constructive criticism that lets him get to a better quality work product in less time.

You're listening to the Agile Attorney Podcast, powered by Agile Attorney Consulting. I'm John Grant, and I help legal professionals of all kinds build practices that are profitable, sustainable, and scalable for themselves and the communities they serve. Ready to become a more Agile Attorney? Let's go.

All right, welcome back everybody. I am really excited to bring on to The Agile Attorney Podcast, Professor Jonah Perlin. And Jonah was gracious enough to have me on his podcast a few years ago now. I think it was, we recorded at Cleo Con, sort of one of those first ones emerging out of the pandemic and it was super fun. So I'm glad to finally be able to return the favor. Welcome.

Jonah: Sure. Thank you so much for having me, John. I don't tell people this often, but whenever someone asks me like, what's your secret, one of your secret favorite episodes of How I Lawyer, your episode comes up time and again as kind of a secret special episode, both because we were able to talk about your path and your path to the law and your path in the law, but also some of our shared topics that we both like to talk about. And it was geared toward law students. So I'm grateful to be on. Thanks for having me.

John: Awesome. No, I'm excited. That's great to hear. Well, let me pitch that right back to you and say, tell me a bit about your path and what you're up to and how you got here?

Jonah: Sure. Yeah. So high level, I grew up in a family that I'm not the son of lawyers. My grandfather was actually a lawyer and a court of appeals judge in the state of Illinois. And that has always sort of been a driving factor for me. But my mom is a rabbi. And so I thought that's what I was going to do and follow in her footsteps.

And it didn't come until college that I actually sort of looked in the mirror and I said, I like many of the things that rabbis do every day. I like helping people. I like speaking in front of a crowd. I like teaching, but I didn't want to do it in that format for the rest of my life. And that's when I started looking at other potential career paths and law jumped out. I wasn't exactly sure what kind of lawyer I was going to be or whether I was going to be a law professor someday was sort of never really in my mind, maybe not at least in the back of my mind.

So fast forward, I go to Georgetown Law School where I now teach. I graduated, I clerked for two different amazing judges, worked as a litigator at a big law firm here in my hometown of Washington DC. And then reached another sort of fork in the road. I knew I liked being a litigator, but I didn't love it and couldn't see myself doing it for the rest of my life, at least at a big law firm. And so I started looking at, what's next?

Luckily, I had adjunct taught a class at Georgetown at night at the worst possible time in my life to do that. We had just had our second kid, so I was like grading papers between midnight feedings. But talk about like finding the signal among the noise. Like I just really enjoyed it and I thought it fit my skill set really well. And I was lucky to find a full-time position at Georgetown, and that's where I've been the last seven years. So, high level, that's me. In addition, how we met, obviously, I started this podcast during the pandemic where I interview lawyers about what they do, why they do it, and how they do it well. I just released episode 143, which is wild for me to think about. And yeah, so that's me in a nutshell.

John: Very cool. And tell me what classes you teach.

Jonah: So I teach in our legal practice program, which for sort of established lawyers, you might know the class as legal research and writing. So I teach the sort of practical class on legal communication for our first-year students. Georgetown has a pretty big 1L class, and so our legal writing classes, despite having 14 full-time faculty members teaching it, are quite large. So I have 60 1Ls this year. They just finished their final exam, so I'm going to dive in and start grading on Monday.

And yeah, so in that class, we teach everything from sort of how to write a basic legal memo to how to communicate effectively as a lawyer, negotiation, oral argument. And I like to say to my students, I get 20 hours with them in the fall in the classroom and 20 in the spring.

So I have one work week to turn them into a lawyer to get ready for their first summer. And whether or not I do that effectively, you'll have to ask them, but it really is a really fun and exciting and I think especially in this moment, important class for people coming into our profession, especially given that so many people coming into our profession now don't have a lot of experience with the profession before they start law school. And that's what makes these classes even more important because they're kind of, they can serve as the great equalizer for those going into practice. So, that's what I teach.

John: Great. Well, and I was going to save this to the end, but I think I'm going to hit it now. Well, the question I'm going to ask, and then I'll actually give a little more background is, what are the sorts of things that you feel that you need to be teaching students today that maybe weren't on your radar when you first started teaching, or certainly weren't on your professor's radar back when you were going through law school? And I'll give you the little bit of backstory. There's a line from Seth Godin that I really like, and I won't get it exactly right because I don't have it in front of me. It's from one of his earlier books. It actually goes back probably 15 years or more.

And he talks about the problem with expertise is that if you start getting your identity wrapped around your expertise, but things change, then it sort of makes you fragile in your ability to navigate the new world. And I think it is something very true with lawyers. Well, it's sort of funny. One of the things I do for fun and punishment is read the various lawyer subreddits on a pretty regular occasion. And there was a post just today about what to do with the 50-year experience lawyer who has recorded three hours of Dictaphone tapes before anyone else even makes it into the office in the morning. Right?

Jonah: Right, right, right, right. Yeah, I mean it's a great question and it's one I think about a lot and frankly, many things about being a law professor, frankly, like being a lawyer, get easier the longer you do them, right? The more classes you've taught, the more reps you have, the more comfort you have, the more willingness you have to take risks, all the important things to sort of being successful.

But I will say seven years into teaching, one of the things I'm lost or losing is every year I'm further and further away from actually having practiced law. And so that in some ways was my, you know, to the extent I had a superpower when I started, I had to learn a lot, but the one thing I had was sort of recent everyday experience. Frankly, it's one of the reasons that I started the podcast was to reconnect me to the profession.

And it's I've been able to build a network sort of practicing lawyers, and not just practicing lawyers who practice law like I did, but practicing lawyers who practice law in many different ways. And I think that's important to stay fresh, at least for the subject that I teach. I really do believe that's important.

But to answer your question a little more directly about sort of the differences that I've seen, and I have an interesting vantage point because 15 years ago I was sitting in the class I now teach in the exact same classroom. Crazy. And look, in that moment, right? There are some from a sort of 10,000 foot view, many of the things have stayed the same, but lots of things have changed.

So when I was learning legal research and writing, it was the first time you really had access to natural language legal research, which is wild to think about, but this was like the change for those who remember it between Westlaw, what they then called Westlaw Classic and Westlaw, I think they started calling it Westlaw Next. And the big difference was you could move from Boolean searching, which I won't talk about, don't want anybody to get too bored, to more Google style natural language searching.

We were just learning how to write emails, like the first time legal writing professors really taught how to write an email, probably is in 2006, seven, eight-ish. And now that's a big part of the class that I teach. So many of the things though are exactly the same. So those are some of the differences, but much of what I teach, how you do basic legal research, legal analysis, argue to rule-based reasoning, analogical reasoning to prior case law, those are exactly the same things that I learned and the exact same things that my professors learned when they took legal research and writing.

John: Yeah. Well, and so that's, you know, near and dear to my heart, right? Because as I work with lawyers and firms and, you know, even a lot of non-lawyer practitioners, I'm huge on this notion that process and technology are not the same thing, right? You have to have a good process, you have to understand what you're trying to accomplish and what are the steps to accomplish that. And then if you can really understand that, then that can make you, you know, lower case a agile or fluid or nimble, whatever you want to say, in order to start adopting different technologies. And I think that's a big part of the evolution from going to the stacks to Boolean searching on Westlaw, to Westlaw Next, to, you know, now this new world of legal Ais.

And, you know, I had Damien Riehl on talking a little bit about Vincent a few weeks ago and things like that. And Damien was interesting because he was a little bit poo pooing the notion of going back to the stacks. Not that, you know, it's not useful to know, but I do think it's important to know the process, right? What is the goal of a legal research task set, for example?

Jonah: Right. I think you're right. I think the process is key. I do not make my students go to the stacks anymore, but we do talk about, right, how law is organized because that helps you do legal research even when you're using online databases. The only thing I'll add though, right, is even if the process stays the same, the technology changes, the other piece that I've seen as a professor over the last seven years is that the base knowledge, experience and ways of interacting culturally or otherwise with the world have changed. And they've even changed in the seven years that I've taught, right?

So I can tell you that there was a pre-COVID sort of there's no such thing as an average student, but there was a pre-COVID average student and there is a post-COVID average student. The students that I'm teaching today, all of them or the vast majority of them either were completing high school or college from their homes, right? And that creates different sets of experiences which inform the people they are today.

And that's in addition to all of the sort of generational differences that we long know exist. And so when I started teaching, the big news was Gen Z goes to law school. I came into teaching right as Gen Z sort of first was eligible based on birth dates to enter law school. And now I'm getting sort of the heart of Gen Z in law school. And you have Gen Z partners now in law firms, very junior partners, but partners in law firms. And so those cultural changes also show up in my classroom a lot about expectations, how people learn, what people know, what people don't know, what people are willing to push back on, and what people aren't willing to push back on.

John: Yes. Yeah, I agree with all of that. And to some extent, it's same as it ever was, right?Technology changes, different generations, and you know, I think of my grandmother and the rubber band ball that she had in her junk drawer because she was born during or, you know, had some of her, wasn't born, she was born before the Great Depression. So I mean, she remembers shortages and need and want. And even, so my family, fortunately, or her family at least, was reasonably well off during the Great Depression. So it wasn't like she even needed to be doing that per se, but it was part of what was in the culture, part of in the zeitgeist. And so she adopted that cultural habit.

Jonah: You know, I think we need to, you know, acculturate junior lawyers into the profession. But people who sort of reflexively fight back on newness, that's just very short sighted in my view. I think we need to sort of meet in the middle. One of my former podcast guests talks about generation Z as the generation of why. They always ask the question why? Why this, why that? And ultimately that's not actually in and of itself a bad thing. It forces us to sort of interrogate our priors. We also were asking why even in ways we didn't know how.

But, right, the flip side of it is they also need to, the junior lawyers who are of this generation also need to be willing to listen to the answer and really interrogate the answer. That's how I try to approach my relationship with my students, is I'm never going to hide the ball and say the answer to this question is because this is just how we do it. I try to make the because more meaningful or as meaningful as I possibly can because I think I owe that to them. But I hope they come in with an open mind and open heart to what the professional expectations are.

John: Yeah, yeah, for sure. Well, and again, I'm laughing a little bit on the inside because in the world of Lean and sort of Lean Six Sigma, the five whys analysis is actually one of the great tools for doing root cause investigations, right? You get, okay, so great, what happened? Great. Why did that happen? And then why, you know, and it often doesn't take you all the way to five. You can get to some pretty meaty, interesting things as long as you're willing to open things up a little bit.

Okay, I'm going to go back to something you said a minute ago and I'm going to make that the segue into the reason that I asked you on to the podcast to begin with, which is you said you are teaching your students specifically how to write emails. One of the books I recommend to people a lot is a book called A World Without Email by Cal Newport. And you and I are both Cal Newport aficionados. A World Without Email is a great book. It is interesting.

I guess even in my practice, I have different firms that have really done a good job of getting off of email to various degrees. Some of them because they've read that book, some of them just on their own accord because they recognize that email is a inferior communication tool to a lot of other things.

But going past that book into the actual segue, you had a post on LinkedIn a month or so ago responding to one of Cal Newport's podcast topics where he really talked a lot about shooting for solutions and things that work at a human level, not necessarily the perfect answer from a technical standpoint or a research standpoint or whatever, but getting to that sort of not just the good enough, but good and also something that people are able to work with, that humans are able to work with. So I'd love for you to reflect or expand on that a little bit.

Jonah: Sure. Yeah. Well, first of all, I can't agree more. I've been a long time Cal Newport fan. I, in fact, even when I was in college and law school was reading his books about how to do well in high school, which is actually how he started. The way back, yeah. Because I'm just curious. I like the way he writes. He happens to teach at Georgetown, although in the computer science department, we've never met, but someday I hope to. And that post on LinkedIn that I wrote was in response to sort of the fact that almost every Monday morning when I'm driving into work, Cal Newport's podcast is my sort of soundtrack for the ride.

And, you know, he and I don't agree on everything and, you know, he takes different tax on a lot of things than I do. But ultimately, what I really what resonated with me in that particular episode, and he's a computer scientist, is he was talking about and reflecting on, I'm going to butcher the science. So if I'm wrong on the science and you're listening to this, just ignore that. He was talking about like passwords as an example, which is that like the optimal password is incredibly complicated, right?The optimal password is one that essentially no one could possibly remember, solve or type in.

But, right, the second that you require one of these optimal passwords, you essentially create huge digital risk because the only way people could possibly type it in is to write it down and basically defeat the whole purpose. And so what he was talking about is the difference in sort of computer science, cyber security, which I've written a little bit about, of sort of optimal approaches and sort of human centered approaches.

And I think that's something we often miss in the law, right? Not just in the law as sort of like how we write laws and interact laws, but I mean as lawyers, how we practice. I think we often create practice-based approaches and process, as we sort of were talking about before, based on what we believe to be optimal outcomes. And we just say we're going to white knuckle it and we're going to get to that optimal outcome as opposed to really thinking what can we as human beings who practice law do?

And I think that cuts across legal communication, legal practice, legal ethics, billing, confidentiality, I mean, basically it touches on every possible piece of what it means to be a lawyer. And it also touches on what it means to be as a lawyer as a human being. I think some of the problem that lawyers tend to have with mental health, and we know that's a huge problem in our profession, is sort of a version of this perfectionist mindset where we set these goals for ourselves even on our processes, processes that are optimal that we could never possibly achieve.

And what does that set us up for? That we've failed before we've ever started. And so that's sort of what I was reflecting on in that moment. So I'm glad you caught it. Not a lot. It was one of my least liked posts, not that I really care, but I sort of as I was reading it, I was like, this is really one of those like deep cut, you really have to like the things Jonah likes in order for this post to resonate.

John: Well, it, yeah, it obviously caught my attention. And there's so many things where that comes into play. And I've probably have mentioned this on my podcast before. It's something I talk about a fair bit. One of, I don't know, my biggest eye roll books in the canon of books for lawyers is the one titled How to Write Bills That Clients Rush to Pay. And the reason why, and I think it's an ABA book, I can remember, it's been around forever. It's got multiple additions. And the reason why that is cringy to me is that if you're waiting to communicate value until the bill arrives two, three, eight, 12 weeks later, you've already lost the thread, right? I mean, you there's nothing you can do to recover at that point.

And I think you're right. I mean, I think that there are so many examples in law practice. I understand where the challenge comes from because one of the things that is true is that our deliverables often have multiple audiences, right? And specifically in litigation, but even in, you know, contracts and everything else, right? We're writing for something that opposing counsel is going to be able to pay attention to. We're writing to communicate something to a judge or maybe eventually to a judge if a contract winds up in a courtroom, right? Or maybe even to a jury, whoever, right, goes that far down. But the party that often gets overlooked when we do all of that writing to different audiences is the client, it's the customer.

And then it feels funny to either number one, to write the document in a way that the client can understand it, because maybe isn't its primary purpose, right? Maybe it really is something that needs to be better understood as a court filing or you're trying to get opposing counsel to adjust their position based on something. I think a lot of lawyers don't do a great job of circling back with the client and saying, okay, here's this thing I did.

And, you know, again, the five whys analysis. There's a lot of root causes for that. I think the billable hour is a culprit where if we're billing the client hourly, we just assume that they don't want to pay for a lay person's explanation, and we're certainly not going to give that explanation for free. At least that's the attitude. I think there are different ways of going about it. Again, different practices can be so different.

But a lot of what I will do for my law firm clients that have at least reasonably repeatable workflows, and that could be estate planning, that could be immigration, that could be criminal defense, that could be family law is we try to come up with a reasonably human readable roadmap document up front, right? And really help sort of give people a high-level overview. And then we can use that as a reference and milepost people along the way. So this thing I just delivered, yeah, you don't have to read it, right? It's designed for this other purpose, but it gets you to this point in your roadmap. And if you have questions about that, then you can, you know, we can go a little deeper.

Jonah: Yeah, I mean, I think that makes complete sense. And it's something, I mean, you know, it warms my legal writing professor heart to hear audience discussed, right? Because the two key things that I will if I told you all the things I try to teach my students, if they remember one thing from the 40 hours that I spend with them, it's always know your purpose and your audience. That is the key to fundamentally good communication. And you're absolutely right.

And I think in the technological age or in whatever version of the technological age in which we live, there are often multiple audiences spoken and unspoken. And the difference between good lawyers and great lawyers is the ones who actually know what audience they're writing for or speaking to at any given moment. That's step one in the process. And it's kind of an analog to what we were talking about in terms of optimal outcome versus human-based outcome.

You're almost certainly not going to write the best will ever or the best contract ever or the best brief ever, but it needs to be a good document for the purpose that you have promised to your audience. That is the yard stick against which one must measure themselves. But if we don't, that's where we get ourselves into trouble.

John: Right, right. I've I'm laughing again only because my former law partner had a funny line he would throw out when we would finish a particular, you know, brief or motion or whatever, and we were reasonably happy with it and he would say, are we the best lawyers ever? Who's going to prove us wrong? How would we know?

Jonah: I love it. I love it. My version of that was the negative version, which is, and my wife would tell you this, anytime I finished an exam or wrote a paper in law school, I would always say, well, it wasn't my magnum opus. Which was my way of saying like, yeah, I know it didn't need to be, but like I know I could do better. Of course, you can always do better. Right?

That's actually one of the things that I think you asked me about generational differences or we talked a little bit about generational differences. I think the current generation is less comfortable receiving feedback. And so it's something that I am really dedicated to teaching and modeling and experimenting with is helping junior lawyers become comfortable with getting feedback and to see feedback as a gift and not a statement about man, this is bad, but actually this is how you can make it better because we can make everything better. So it all comes back to that same idea.

John: Well, and that's interesting. So that's a topic that I'm actually really passionate about because I think, I mean again, in the world of Agile and Lean, feedback loops are a big part of the practice set, right? The tool set that we use and you don't get improvement, you don't validate your assumptions about things without feedback loops. And so I think there's a few different layers where your comment comes in.

One is I think it's important for law firm owners and law practice leads, etc., to create a culture of feedback, of healthy feedback. But I think another challenge is certainly, right, I'm a solid Gen Xer, which means that I learned largely from boomers and even, you know, greatest generation folks, the pre-boomers and they weren't always the best at giving feedback, right? They're good at pointing out flaws, but that's not the same thing as giving feedback. And so I think that there is learning that needs to happen on both sides. I think I have talked about this before. One of the great, I coached youth soccer for about 10, well, 11 years here in Portland, despite never having really played much organized soccer in my life.

And one of the early pieces of feedback I got was from an opposing coach. And he said, hey coach, can I offer you some feedback? And I said, yes, please, of course, because I really truly didn't know what I was doing. He said, you're remote control coaching your kids. I was like, well, what do you mean? He's like, you're just trying to basically, you know, say, hey, Jimmy, go over there, kick the ball, pass it to Joe, etc., etc. And he said, think about and I get I think he'd gotten this feedback from someone else too. He said, don't tell them what to do, help them see what you see.

And that was really like groundbreaking. I mean, it changed my whole approach to how I was working with kids and frankly, even working with adults, right? I really I think it's so powerful. It's so powerful. And another way of saying that, and the way that I often translate it to my consulting clients is we have to help understand the why, right? You can't just say what needs to happen. You know, certainly you want to have an exemplar, right? What should this look like when you're done, right? That's good pedagogy in terms of teaching new things.

And then having something of a recipe, which is, you know, a formula, a process, a system that says, okay, in order to achieve an output like this, you probably should roughly follow these steps. And again, that is important, but neither of those two things is as important as helping someone understand what's the purpose and to a point you made a minute ago, who's the audience for this particular piece of work and then maybe even deeper still, how it fits into the bigger picture of the goals of the client, of the firm, of the business, etc., so that you can really lock it in and kind of understand the context.

And part of the reason why that's so important is because then when people make the inevitable mistakes that people make, they will at least tend to be falling in the right direction.

Jonah: I like that. Yeah, I mean, I think that's absolutely right. I think it's so important and I think you're absolutely right that we need to train junior folks how to receive and learn from feedback. But if it takes two to tango here, they are not the dancer to extend the metaphor that's leading. The person who's giving the feedback really needs to be critical about how they do that in the best sense of that word. And we don't train lawyers how to give feedback either.

And so that's something that we have to model and I think any firm lawyer that tells me, oh, this generation can't write and they can't learn from feedback. My next question is going to be, well, how do you give it to them? And they say, oh, well, you know, I spend a lot of time giving them detailed red line edits. I'm like, great. Do you explain why you make those edits? Do you give them time to sit with those edits? Do you have a post filing meeting? Well, no, then there's no surprise and I'm not judging them for it. I want to make it clear that they have not prioritized it. So therefore they should expect that outcome.

John: Well, and so this gets back to, I think the thing because I went back and I hadn't listened to that Cal Newport podcast episode until you mentioned it, but I did listen to it. And it's this idea that doing the thing that might feel inefficient in the short term is actually hugely more efficient or more productive, you might say, or I might say, in the long term. And so using that feedback example, I have several clients that have transitioned to when they do red lines, they turn on a loom video and they do red lines while doing a screen recording and they're talking into their microphone as they make changes. And so then instead of just sending a red line document back to their team member, they're sending the red line document plus the screen recorder.

And that's not quite as good as bringing them in and sitting down and doing it together, right? Which would be maybe the most optimal way, but it's hard to coordinate schedules and all the rest. But doing it into the screen recorder, aside from the fact that you have to sort of talk through your process, it doesn't really take that much longer.

Jonah: Right, right. It's it's an it's an efficiency hack. It's a it's a thoughtful way of doing it.

John: And it gets people and the attorneys, the senior attorneys that I work with who are doing that report that their juniors, their paralegals, their associates, whoever, really like that approach. It really helps them connect with they feel like they're getting personal attention. And they also feel like it's less critical, I think at that point, right? It really does wind up being more supportive.

Jonah: Yeah, I mean, for anybody listening who's never tried it, I will tell you, I have had, whatever it is, several hundred students over the last seven years. I have never given one red line to one student ever. That is not because I don't spend most of my days reading student work and giving feedback. We give all what I call comment bubble feedback, right? Which means I need to explain what I want them to change. I might include some words that I think are suggested ways of changing it.

But that process, which is far more time consuming on both the feedback giver and the feedback receiver is I think the only or maybe not the only way, but certainly the best way that I know how of giving asynchronous written feedback on legal writing. And the other big tip is just force yourself to use the word because until you're blue in the face because that is what moves you from this is how I would do it to as you said, sort of showing them how you think and building that into their practice.

John: I love it. Yeah. Okay, great. So we've got a few tips that maybe people have heard already from me, but a new one from Jonah. And I think that part of it is, and again, I don't know if this is quite the perfect analogy, but I really do try to encourage the law firm owners, the senior partners that I work with to think of themselves as educators. And I don't think that's necessarily the self image that a lot of senior lawyers have certainly traditionally if had. I mean, there've always been great. You always hear these stories of like, oh, I learned so much from this person.

So I think there's always been sort of natural teachers, natural educators that are also very good lawyers and wind up being able to teach it. But I think a lot of people maybe don't always immediately draw that connection to what they're doing in terms of again, building their firm, building their team.

Jonah: Yeah, it's an echo back to the apprenticeship model where our profession started, right? The only way to build new lawyers before we had law schools and frankly, despite working for one, the only way we can build new lawyers despite having law schools is for people to continuously train people who are a little less experienced than they are. And that takes commitment on both sides of the equation.

John: Yeah. Yep. I think that's right. A couple weeks ago I had Carolyn Elefant on the podcast and one of the things that she talked about, we talked about how it's a great time to start your own law practice. And one of the points she made is that the marketplace of contract services is also better than it's ever been, right? It's really robust.

And you can just sort of buy certain solutions. And I think that is correct, but I think that maybe we take that to an extreme or some lawyers and business owners take that to an extreme and think that, oh, there's a market for everything. And whether there is, whether there isn't, sometimes you're just going to get better results if you build it rather than buy it.

Jonah: Mhm. Absolutely. And even if you buy it, you'll be able to assess it better if you know how to build it. I mean This is this is the you know, if you were to be a fly on a wall of any conversation between two legal writing professionals right now. This is the conversation we're having about how do we teach in the age of generative AI?

I have lots of ideas but no solid answers, but it does come back to sort of that foundational principle that you need to know how to do these core skills regardless of what the task that you're specifically asked to do is. I assume the task will change, but I don't think, at least now, the core skills that get you competent to do that task well have changed.

John: I think that's right. Yeah. Although that brings back to mind a thought I had while you're talking a couple minutes ago, one of the great uses for generative AI is if you do have to write a legal document that is specific to a targeted audience, you can feed that into a paid version of GPT or Claude or Vincent or whatever and say, okay, great. Now please explain this document to my client and do it at a 10th grade reading level or an eighth grade reading level or whatever you need to do. It won't necessarily give you a an email ready version of it, but it'll at least give you some bullet points you can work with in order to start that communication.

Jonah: Yeah, I'll just tell you a cool version of that from my end. So in addition to teaching, part of my job is writing scholarship and I'm working on an article right now about generative AI and its interaction with the billable hour. And I was working on the introduction just this morning and I was using a tool called lex.page. It's basically just like a word processor that has a GPT chat built in. It's a very interesting tool and I was sort of playing around with drafting my introduction. And I'm one of these writers who writes and writes and rewrites and rewrites and I've been working on this introduction literally for weeks and weeks and weeks, but I'm close to sort of the version that I'm ready to share with some trusted colleagues.

And the way this tool works is you can actually highlight text and ask it a question sort of in line and it can read the text and bring the text sort of in the back end into the prompt. And I asked it to give me feedback on the introduction or parts of the introduction as if I were, and then I put in the names of several specific sort of leaders in the field of legal ethics. And I went through them one at a time because I sort of I've read a lot of their work. I know what their interests are, what they've written about. I did not feed their articles into the LLM at all.

And it actually gave me like different feedback for each of the professors names that I used because I guess they've written enough that it was built into the model itself. And it was so helpful because that's one of the audiences I'm writing to and it's the hardest audience for me to recreate without actually giving them my early draft. And it was extremely helpful in figuring out sort of what stays in and what gets cut out, which is sort of the hardest part, I think of a law review article introduction. It was an incredibly eye opening experience and one I will definitely do again.

John: Yeah, that's fascinating because I would venture that exercise would be important and useful regardless of how close the AI actually under right? Because of course it doesn't understand in the classic sense.

Jonah: No, it doesn't understand anything.

John: Right. But what's important, it's it's a little bit like the oblique strategies deck. Do you know this from Brian Eno? Okay. So Brian Eno, famous musician, producer. There's a great documentary that you can dial up for him. Although one of the things about the documentary is that they filmed it in a way that it's modularized and when they showed it in the theater, every time you saw it, you saw a different recombination. So you could actually go see it several times and it was a different thing each time. This is a very Brian Eno thing to do once you understand him.

But the oblique strategies deck was something he developed with somebody else, Peter Schmidt is his name. And when they would go into the music studio, bands would get stuck, right? Humans get stuck. And he came up with this deck of printed cards and when they would get stuck, the instruction would be to just draw a card and respond to it. And I'm drawing one right now. What are you really thinking about just now incorporate it?

Right? And it's just enough to sort of when your brain and when your anxiety or your worry or your doubt or all of these natural human things click in, it allows you to just sort of chink off of that track and get out of whatever rut you're in and sort of break you into a different way of thinking. And I find generative AI is tremendous at that in my writing as well.

I don't go with it for writing, but I have it give me feedback because it's basically cheap feedback and eventually quality feedback matters. And I think that the quality of feedback from these tools is improving all the time. But any sort of feedback is useful. And there might even be a value in having it be a non-human because there's just no emotion, right? Or there's no judgment in it, right? I'm not going to feel bad if chat GPT or Claude or any of these tools quote unquote criticizes me because whatever, like what's it going to go? Go tell my mom?

Jonah: I think that is exactly right. And I think it is such a better use of the tool, at least as currently constructed than write this paragraph in the style of. This is give me feedback in the style of. And these are tools that are predicated on pattern recognition. That is their bread and butter. And so I was really impressed with the unique nature of each of the feedback. And because I know these folks writing, I was able to sort of, right, we talked about like being able to be able to verify it. Like I'm not sure that if I actually sat down with these folks who some of whom I now know personally, that's what they would say, but I absolutely am 100% sure that their writing is what inspired the specific feedback it gave me. Like it was accurate in that sense of the word accurate.

John: I love it. Yeah. And I do think, right, this is one of the challenges and I've got teenage boys right now who are in high school and learning to write and obviously learning to write very differently than I did. And also their teachers are try having to navigate and figure out what to do in the AI world. And I think mostly are avoiding it so far, which is an approach. It's okay, right? I mean, I think we're not going to be able to be that way forever.

But one of the key things is that we all have to come up with our own sense of self and style and what is the way that we write? Because writing improves our thinking. And, you know, it's another random thing I'll bring up, but there's a book by a professor named William Zinser, who I think has passed away, but it's, you know, it's kind of a classic from the 80s or 90s called Writing to Learn.

And I really do think that there's value, right? It's another one of these examples of slow productivity, right? You can't shortcut the process, right? You have to actually go through the exercise of wrestling it with it yourself so that you actually know what you want to say because you don't always know what you want to say when you first put the pen to paper.

Jonah: Right. And that's the great part is what the tools allow us to do now, right? Is it allows us to go through the same process, which is I need to write to think, but it allows us to test ourselves or rethink or rejigger in so many different new ways that, you know, it's why one of the things that I'm thinking a lot about is there's this assumption that I think has some truth to it that by definition, using generative AI for legal tasks is going to make many of those tasks faster than they were without generative AI. I think that's probably true in many cases.

John: I think it already is true in a lot of cases.

Jonah: Right. But, but I think there is a world that especially for the highest quality documents and the most valuable creations that lawyers have to make, that they might take the same amount of time to create, but their quality will actually be better. And so it's not a time savings so much as it is a tool to make it better and faster. And the analog to go back to what we talked about is legal research, right?

Before you had the ability to call up every single case that's ever talked about a particular point, whether or not it said it was talking about it or not from your handheld device at your child's soccer game, right? Allows a whole set of workflows that are not possible when you have to go through pocket parts, which anybody who's younger than me and listening to this is not even going to know what I'm talking about.

John: Right. That by the way, my very first job ever was updating the pocket parts in my uncle's law firm. So but I might have been the last person through law school that had actually physically done that.

Jonah: Yep. Yep. Yep. So I hear you and I'm with you on that. So it's a journey right now for sure.

John: Yeah. Well, I think I had another topic, but I think we're running low on time. So I think I'm going to leave it at that except I will just echo the point that I actually believe that all technology, even though we think that we are trying to get and use the technology in order to make us faster, more efficient, whatever, the actual benefit always comes from quality improvement.

And quality improvement could mean a faster way to get to the same quality standard or it could mean an increase in the level of quality. But either way, if we think about all of the work we do through that lens of improving the quality, the efficiency is going to be a happy side effect. Well, Jonah, thanks so much for coming on. This has been super fun. Maybe I'll have to have you back and we can talk about the legal ethics stuff that I was going to ask you about.

Jonah: Happy to do it. Happy to do it. As I said, I'm a fan of your work and it's been fun to just spend a Friday afternoon talking about things that we both love. One of the coolest parts I think about podcasting in general and sort of social media in general because there are a lot of bad things about it that we see every single day is the ability to create connections with people who have similar interests across time and space in ways that frankly were impossible even 10, 15, 20 years ago. So I'm grateful that the world put us together and that we got to have this conversation.

John: Absolutely. Me too. All right. Well, thank you and enjoy your weekend.

Jonah: Yeah, you too, be well.

John: Alright. There was a lot in that episode, but two key takeaways I hope you'll take away from this conversation. Number one: every effective legal communication starts with understanding your audience and your purpose. And it's a simple idea, but it's one that makes a huge difference in clarity and impact. And when you're writing to multiple audiences, it can be a great use of generative AI to help you craft slightly different versions for each specific end user.

Number two, writing isn't just a way to explain your thinking, it's often how you get to your best thinking. And that's consistent with this Cal Newport notion of slow productivity. Getting things right is more important than just getting them done quickly because the rework and failure demand that comes from lower quality output is just gonna take up more time in the long run.

That said, you can use generative AI to accelerate your feedback loops and make sure that your output is actually reaching the audience that it's intended for. Okay. That's it for today.

If you found this episode interesting or useful, please share it with someone who you think would benefit from a more agile approach to their legal practice. And if you want to explore more effective ways to work in your own law practice, we should chat. You can set up a Zoom call with me through my website or shoot me an email at john.grant@agileattorney.com, and we'll set something up.

This podcast gets production support from the fantastic team at Digital Freedom Productions, and our theme song is Hello by Lunara. Thanks for listening, and I'll catch you next week.