Justin True: From Prompt Panic to Practical AI Workflows for Lawyers

Voiceover: Welcome to the 1958 Lawyer, the show for attorneys who know the old model isn’t working anymore. The billable hour, the 12-hour day, the expensive office no one visits. Your hosts, Ron and Rebecca Bockstahler, are here to question all of that and more by exploring smarter, healthier, more flexible ways to run a law firm. No more burnout, no more outdated expectations. Just real conversations with attorneys who are proving there’s a better way. If you’re ready to build a law firm that’s profitable and livable, you are in the right place. Now, here are your hosts, Ron and Rebecca.

Ron Bockstahler: Welcome to the 1958 lawyer, the podcast about the business of law. I’m your host Ron Bockstahler, joined by my co-host RB. Three years ago, two New York lawyers were sanctioned for filing a brief full of cases that ChatGPT simply made up. Since then, courts across the country have kept handing out sanctions. One recent penalty topped $100,000. Every solo attorney is being told AI will save them hours, but almost nobody is handing the guardrails. And here in Illinois, the Supreme Court and the ARDC now have AI guidance you’re expected to know. Our guest today, Justin True, an AI and marketing leader at JDT Advisors who spent years deploying AI inside growing businesses and who now helps lawyers do the same thing ethically and without putting their license at risk. Justin, welcome to the show.

Justin True: Thanks, Ron.

Ron Bockstahler: Good to have you. I’m kind of excited about this. It’s a fun topic.

Justin True: It is a fun topic.

Ron Bockstahler: Tell us how you got started because you started a marketing company, and then you’ve transitioned into AI.

Justin True: How do you how do you become an expert at something that’s only been around for three years? It actually started when I was 10. I learned how to code HTML, kind of in a coincidental way, and that really pushed me into technology at an early age. I was into Photoshop at like 1314, Downloaded illegally off a Napster. I was in marketing and technology before I knew marketing and technology was a field. Inevitably, I went the marketing route, but always been surrounded by technology my entire professional career. I was I was one of the first million users of ChatGPT when it came out in 2022, and was just instantly obsessed and immersed. And over the first year or two, that was right when also when I started my marketing, my fractional marketing business, and I was able to put off hiring a team of support staff because I was becoming this Superman overnight. I mean, there was 2220, 23. I was starting to finish my work that I had deliverables to clients, and I would just like put it on the side and set a timer for three days because I didn’t want them to know, you know, what my superpower was. And yeah, then then I started just get so deep in it, and I had to actually start like refraining myself from using the tools so much because I had a job to do, and it wasn’t AI. It wasn’t to dig deeper. It wasn’t to learn. I had I had a career, a business to to keep up. And about a year ago, really just happen chance, I started to get pulled into AI directions with with clients and with some communities I’m involved with, and I saw an opportunity to take a hobby and turn it into a a full time full time job, and it was it was it was a it was a bet, it was a risk, it was a gamble, but it went yeah it was a full house.

Ron Bockstahler: But you you chose law firms because you you work with other firms, but you kind of really have a real niche within the law legal industry, and there’s a fine line between saving time using AI and getting sanctioned. So, kind of walk us through what you’re doing and how you’re helping these attorneys.

Justin True: Totally. Well, I chose law because a lot of my community was was lawyers. I’ve I’ve had experience with on the marketing side, but also I saw the legal sector as the most data sensitive and secure collection of of people in an industry, and I thought if we could if we could crack this, and it would make working with everything else a lot easier. Now, to that point of of saving time, saving money, all that stuff, like AI is so vast and so big, I really try to get people to not start with the sensitive, complex, ultra secure data, and that. And if you if you jump the gun there, you’re you’re really you’re you’re now starting at level 10, right? And there’s a lot of levels before that that you got to earn your keep and work through to to know how to do that.

Rebecca Bockstahler: So then, and we’re going to talk more about those things, but maybe we can bring it back to how it all began: Mada versus Abianka with law firms, two attorneys, one brief written by ChatGTP, and attorneys got sanctioned. Can you walk us through kind of what happened there, what you might know about that, and then why is that something that is still. Three years later, talked about and totally,

Justin True: totally, yeah, high level. It was I need you know similar cases to to have evidence to support this this charge or this direction, and ChatGPT hallucinated. It made up the answer, and you know this is this is in the early days, right? You know, I I believe that that you remember the McDonald’s coffee cup situation. Yeah,

Ron Bockstahler: that’s when I got burnt.

Justin True: Yeah, yeah,

Rebecca Bockstahler: we’re both. You know, and it like

Justin True: wasn’t even that hot, and you know, there was just it got heard by everybody, and it really made the image you know heard around the world. And I I just see that that’s the same thing here. I don’t think a lot of people, especially in the legal space, even knew what hallucination was until that happened. I mean, if you if you if you go look up something in a book, you know, 30 years ago, 40 years ago, you’re doing a report. You’re probably going to like cite your source, and I think in that situation, the the attorney didn’t go and do their extra the homework that they’re supposed to do. And you know, would that have happened anyways? Absolutely. Would we all know what the word hallucination means today? We still would. It’s just that was the one, the the shot heard around the world, and I think a lot. It scared a lot of people. It scared a lot of attorneys. It gave a lot of attorneys an excuse to not use the tools. And you know, I think it was towards the end of last year was when I started to see all of the mega firms around the world publicly announce that they were using Harvey Lagora or even some of the general tools, and it was overnight. I started to see attorneys all over the place just like put their heads down because the the same 14 reasons why they weren’t using AI were no longer a reason that they couldn’t use AI. Well, you

Ron Bockstahler: go back to that case. It’s interesting. The judge didn’t sanction the attorneys for using AI. He sanctioned them for not verifying the material in the in the brief,

Justin True: which they would have to do.

Ron Bockstahler: Well, I mean that’s the same as if you had a paralegal write in the brief and you didn’t go review the work, right? So I think they

Rebecca Bockstahler: reviewed it by asking ChatGTP if it was correct, and ChatGTP said, “Of course I am. I think they

Ron Bockstahler: doubled down because when the judge and the opposing counsel said we can’t find these, produce them, they went back to ChatGPT. Chat, you know, kept lying, kept hallucinating, saying, “Well, here this is, and you know, more hallucinogen. So they doubled down on their mistake and then didn’t go verify it, and that’s what they got sanctioned for.

Justin True: Totally.

Ron Bockstahler: So now, do you go against? I mean, the ABA has basically said you can use ChatGPT. So is this a big part of your conversations with your clients?

Justin True: Yeah, I think it was. I think you start to you realize a lawyer, a law firm, they learn that there’s now rules to the game, and they they learn them, and they kind of keep them at the forefront, and know that they update. You know, letting the world know that there’s now rules was kind of a one time thing. Now everybody knows. Now everybody tries to keep up and keep track and keep score, and every state has slight variations and differences. And so, if you’re working across states, then got to be aware of you know just same, just like all the other rest of the rules.

Ron Bockstahler: I just think you know. I think we read that Kirkland Ellis invests a half a billion dollars. Yes, they do. Is that a wake up call to say I got to start using this if I’m going to compete with the biggest firms? If the biggest firm just invests a half a billion dollars, they must begin to they’re going to use it.

Justin True: Yeah, they that that was a there’s a very very very big bold strategic bet that they placed. Their real rationale and reason was they they were spending a ton of money on on Harvey and Lagora and all these other tools. That those tools were were you know I don’t want to get confused with training the models, but those tools were learning how to operate in the legal space, very specific off of one of the largest law firms. And Kirkland saw that that was well. This is our information, so let’s just build our own.

Ron Bockstahler: Right.

Justin True: I mean, I think that that price tag nobody can afford that, not many. Right. So they’re really placing a long game bet, and honestly, I I think you’re going to start seeing that across lots of industries. You already are seeing that, yeah. Even in even in smaller spaces, just with the cost of these tools rising, we are entering a world where build is becoming cheaper than buy, which was not accurate, was not true even a year ago.

Ron Bockstahler: We, you and I, have talked a lot about AI, but I want you to walk me through when you’re talking about they’re going to go build their own environment, their own sandbox. They’re building it within ChatGPT or Claude or within some existing platform. Yes or no?

Justin True: No.

Ron Bockstahler: Okay.

Justin True: No, they’re they’re they’re doing big boy stuff. They’ve got massive teams, massive scalability, massive technological understanding to build their own environment and pull the tools that they need inside of it. Meaning they can pull a ChatGPT, they can pull a Claude, they can you know. There’s there’s there’s hundreds of models out there,

Ron Bockstahler: right?

Justin True: Average person can name like three, maybe. So having building a secure environment in a place where you can bring everything inside allows you to build your own your own building and put whatever you want inside of it. What we’re what we’re moving towards is the ability for that to become attainable to smaller firms, and you can do it without having to go that full route of of having your own massive teams, you are going to get it at a smaller scale. But ultimately, we are all moving towards a place where you have a you have a closed container where your information, your data gets stored, models get processed, and you’re in a little lone lock box so that you can confidently say to a judge or a client, that’s where this is being processed and how and by what tool. This is going to become normal in the next year.

Ron Bockstahler: So this moves me. I’m kind of moving ahead. Sorry, RB, but I want to know about security. Talking about if I’m building my own model, how smart’s my model? Where’s it? Is it where’s it getting its knowledge from?

Justin True: You’re not going to build your own model.

Ron Bockstahler: Okay,

Justin True: you’re gonna you’re gonna build your own your own black box that you’re gonna pull a model inside of.

Ron Bockstahler: Okay, so

Justin True: that information can’t leave that room. Building your own model is what Kirkland and Els are

Ron Bockstahler: okay

Justin True: doing. That that it’s extremely expensive and extremely technical to to really build your own model. What what we’re what I’m what I’m referencing is is bringing existing models, whether they’re the top tier ones that you have heard of, whether those whether they’re the very competitive ones that you haven’t heard of, that are cheaper and more expensive, and bringing them into your closed container so that you can run cutting edge technology on your own data,

Ron Bockstahler: are you helping these firms create their black box?

Justin True: Help them navigate. I don’t build. I have I have team and resources that do build that I work with when it gets to that level. But frankly, even your firms up to 100 people don’t necessarily need to build really crazy intense things because it’s all about starting at an appropriate level. Like you got to start at step one. Have to start using

Rebecca Bockstahler: it.

Justin True: You have to start using the tools that are out of the box capable and safe to use for safe and and logical things. You know, don’t put don’t put PII into public tools. But can you write can you write a blog with it? Sure.

Rebecca Bockstahler: So then, if we bring this back to everyday use for small or solo firms right now in Illinois, the Supreme Court put out a policy, the Illinois Supreme Court, and then the ARDC put out a guide of how to use AI, which I think is sort of ahead of what some states are are doing, I believe. Anyway, so based on those two documents, if you’re an attorney right now using AI, what do you need to make sure you’re following and doing to be using it correctly? I

Justin True: actually don’t know the the one two threes to that answer.

Rebecca Bockstahler: Okay,

Justin True: yeah,

Rebecca Bockstahler: that’s fair.

Justin True: I think the place to start though is with non-legal private information.

Rebecca Bockstahler: Yeah.

Justin True: Start to get to use, start to get to know the tools and how to how to how to put public information and how to how to ask questions, how to learn to put information in and get information out, and and discover how to change the shape of that information along its journey, because you can have a super black box safe, but if you don’t know how to use it, then why’d you just why why why’d you buy all that money? And so go play in the playground, and let’s let’s climb the monkey bars. And then when you’re when when you can make the whole stretch, then let’s talk about starting to bring client data. And it all depends on what type of law too, because there’s all sorts of different tools that you can just get that are legal specific focused out of the box that can solve specific needs.

Rebecca Bockstahler: Do you just spend your days watching people’s minds go just blown constantly? Yeah, it’s a lot.

Justin True: It’s a. I like to interact with people.

Rebecca Bockstahler: Good thing.

Justin True: And so yeah, it’s it’s a it kind of keeps me going. Yeah,

Ron Bockstahler: bring us back to the basics. If we’re going to get started, I’m a solo practitioner. I’m thinking, you know, AI’S powerful. I want to use it. What maybe one or two things I should start using it with to get used to

Justin True: it? Totally. The two on the non legal side of the house. Think about: Do I need to write blog content? Do I want to update the content on my website? Do I want to make posts on social media? Can these tools help me figure out a narrative? I want to do a a short video clip, 32nd video clip. What is this topic going to be about? The tool. Then you tell the tool, ask me about myself and my type of law, so that the tool will then interview interview you and help you navigate what could be a good 32nd script for a video. Obviously, that video is going to be public, and so we’re, you know, it’s okay to talk to AI about something like that. So, so start by thinking about all of. The non-legal aspects of your small law firm, small business, because that’s where you get to play, and it’s okay if you fall and scrape your knee. Get back up; nobody’s going to see. Once you are starting to comfortable the tools, start to think about the parts of your legal practice that you’re going to be doing public research anyways without using your client’s data. Hey, what are the statures on on this law, this law, this state? Can you can you find me the most recent regulations on this type of topic and this type of industry? Start using the tool as your research powerhouse assistant for public information, and it’s just taking you to that next level of how to interact with these tools, and when you’re ready, when you when you say to yourself enough times, be really helpful if I could put my client data in here, like I could really take this to the next level. That’s when we introduce you to some of these off-the-shelf tools that are black box that will just take you to that next step of same research, same methodology, but I’m now able to pair it with some sensitive data.

Rebecca Bockstahler: So, can you kind of maybe bring me into like this whole black box situation? Is it a setting that people need to know? Is it a development that they’re hiring someone to do for them. How how does the black box part happen?

Justin True: Different different levels. Okay, as you probably assume. I hear a lot of a lot of like don’t let the model don’t train the model with your data. You’ve heard that.

Ron Bockstahler: Oh yeah,

Justin True: that’s just one of like five layers. Yes, you don’t want the models to be learning from your information, but your information, you can turn that setting off to, and I’ll continue answering your question. You can turn that setting off. It’s the basic setting of don’t train the new model on my data. Okay, but your data by being put into the tool, your data, your file is now sitting on the servers of other companies. So it’s not just about not training the model; it’s about not letting your data sit on a major corporation server that you don’t know about. Which was when we start to get into data retention standards and how long can a company hold on to that file? It’s not until you start to upgrade to the to the enterprise, the much more expensive versions of these, where you can start to control how long they have your data. And it’s when you get to theirs when it’s okay to start to use these big boy tools with these big boy price tags, but to to answer your question, a black box though is a little bit more expensive than what you’re gonna pay 20 bucks a month for. You’re you’re gonna pay a couple of $100 a month for, but you can have an off the shelf box that is requires zero setup on your end. Okay, you get a new login into a very secure container, and you get to use the same tools that you hear of and that we all talk about, but they’re in a different container that’s safe, and it’s this shortcut to having law firms start to use their sensitive data without having to go the Kirkland L’s route,

Ron Bockstahler: let’s step back maybe one step because I want it can be. Let’s be basic. We what you were referring to earlier was prompts. How do you write effective prompts? How do you interview me? Right. I mean, you can go to your your AI bot and say, act as my controller, act as My CFO,

Justin True: yeah. Let’s talk

Ron Bockstahler: about basics of writing super script.

Justin True: So right around the time that what is it, hallucinations came out. That that word came out, as did prompt engineering. Probably have heard that the art of writing a good prompt, and it was a it was a buzzword for a couple of years until it wasn’t a thing anymore. You don’t have to be good at writing a prompt. When I work with people, I refuse to give them booklets and packets and lists of prompts because it’s counterintuitive. Because how you need to interact with the tool, even if we might need the same thing. It’s going to be different how I interact with the tool. What you need to be good at instead of trying to convince yourself to get better at writing a prompts, you need to convince yourself to be better at communicating. If you have an intern, a brand new intern shows up today, you have no idea their skill level, what they’re capable of, what their history is. Are you going to say, “Go write this report for me, and then then you’re going to expect genius to come back, or are you going to be very detailed? Are you going to type a long requirements document to him, and are you going to judge yourself on how you’re writing that? No, you’re probably just going to go up to him, and this I’m saying this because my first. First number one tip is to turn the microphone on and just start talking to the computer. We all sit in front of Zooms all day long talking at a computer. Get used to

Ron Bockstahler: it.

Justin True: When you open up a tool, turn the microphone on and just start talking, as you would to this intern that showed up today. Hey, I need you to work on this project, and here’s what I want you to accomplish. But before you get started, ask me five questions so that you can understand the context, and I can understand that you know your job. You’re not. We’re not. We’re not. We’re not crafting prompts. We’re communicating. Right. I’ve never

Rebecca Bockstahler: used my microphone with these tools before.

Justin True: Will change your life. Head explode. Exactly. Speechless, really

Rebecca Bockstahler: bad during a podcast.

Ron Bockstahler: Let’s go back to you. You mentioned blogs, and I still talk to a lot of attorneys. Like, well, it’s not coming from me, and and it’s a conversation that it is coming from you. If you’re the ones interacting with the AI, the AI is assisting you and doing the research to create that blog. Can you talk about because you started this many many years ago? You did it when you were young. You know you set it aside. It’s an effective way to get good information out there, and and you’re getting the the credibility for it. Like your if it’s on your website, you’re getting your authority scores going higher. Let’s talk a little bit about that.

Justin True: Totally, Assuming that these tools are just writing your blog is is jumping the gun. It’s it’s making it’s it’s it’s it’s it’s assuming that nobody’s putting any other work or thought into it. That you’re just saying make me a blog, and you again we just talked about this intern, right? If you were having an intern write a blog, you’d probably give them a little bit of direction on it. They’d probably write you a draft. You’d probably tear it to shreds, and you’d keep going round and round. Well, with these tools, you can you can just expedite that right all the time. I listen to podcasts all the time on all sorts of topics, mostly AI. And when I hear a topic that I like that I resonate with, I’ll go grab that transcript of that five minutes or two minutes or whatever. I’ll drop it into my tool and I’ll say, “Hey, ask me five questions to get my opinion on this. And guess what? I’ll turn the microphone on and I’ll just talk myself through all of these answers. And so I’m now getting all of my, I guess it’s rich thought into the tool, and then I’m having the tool organize it and massage it and smooth it and get rid of all the ums and the this and the that’s and the the trails that I go on and that’s my blog article

Rebecca Bockstahler: you

Ron Bockstahler: right

Justin True: I just didn’t write it because I don’t I don’t do that.

Ron Bockstahler: I don’t have to.

Justin True: I don’t have to. I also don’t spell well. I’ll spend more time trying to see through all the red lines than getting my actual thought out.

Ron Bockstahler: It’s it’s it’s when pandemic hit, we all jumped on Zoom. We all started having these remote meetings, and I became smarter. And it wasn’t that anything’s really changed except when I was talking to you. I could sit. I’m sitting at a computer. I can look things up, and when my memory, which isn’t very good, especially for short-term things, and I’m like, I know what it is, but if on my computer, boom, I could pop that up, and I’m having a conversation, and you didn’t know I just did that, but now my memory was so much better. Of course, you meet me in person. Like, why can’t that guy remember shit? Who is that? The same guy. But it’s kind of the same thing now. Is I mean, I sit and write a blog, and it could take me two hours because it gets me asking questions and thinking. So the research behind it, in my opinion, is way better than if I was just sitting hand writing something off my memory.

Justin True: Totally, they really get to be an extension of your brain. Absolutely, access to basically all public information, and if you know how to tap into that,

Ron Bockstahler: yeah, the

Justin True: world is limitless.

Ron Bockstahler: And I think that’s what people got to realize is when you’re writing, you’re also writing and doing the research and incorporating that into your article while you’re doing it, but still takes. I mean, it still takes me a couple hours to put something quality.

Justin True: Absolutely.

Ron Bockstahler: And the far deeper I get into it, longer it takes me.

Justin True: the The quality of my output of my written report outputs that I do for for clients. Well, they tell me it’s nice. They tell me it’s awesome. I know that I would not have gone to these depths and of that quality four years ago. Right. I’m doing things that I never would have done or could have done. And at a at a holistic macro level, I love this concept. There’s this is not just about law. This is about every. There’s the world. There’s two types of AI. There’s efficiency AI, which is where you’re doing the same with less, and then there’s opportunity AI, where you’re doing more with the same, and more meaning things you never did before, either because you were. Couldn’t have, or you couldn’t have, and when everybody starts tapping into opportunity AI, you start to have a world full of stuff that didn’t exist five years ago. That exists both at the micro and the macro level. So it’s it’s my biggest goal is to help people and their businesses along this journey because there’s a lot of opportunity out there to do some really cool things.

Ron Bockstahler: It’s gonna, I think, it’s gonna create work. I’m on that aside where it’s gonna create more more opportunities for people as they’re going along. Because what you just said is the opportunity AI allows you to do so much more. So our

Justin True: our everybody’s job is made up of a number of tasks. Usually, there’s a lot of tasks wrapped up in one job. Is AI taking jobs? Not really. It’s taking tasks.

Ron Bockstahler: Yeah. And

Justin True: when you can take one task away from one job out of when that job is made up of dozens or hundreds of tasks, you’re creating room now for other tasks to take up more meaningful time, or you’re taking a lot of tasks. And what do you think? A person’s just going to lose like a quarter of their job? No, they’re going to figure out something to do with

Ron Bockstahler: it.

Rebecca Bockstahler: Right. I guess it just goes back to what we always say: if you have to learn and adapt the tools, because it’s not that the job gets taken; it’s that you maybe get left behind.

Justin True: Did did computers really wipe out anything? No, they sped it

Rebecca Bockstahler: up.

Justin True: Did autopilot get rid of pilots? No, it made it made it made a surge of pilots.

Ron Bockstahler: Let me add to that one. We we sailed our our boat over to St. Joe’s this past weekend, and autopilot wasn’t working. We had to get it replaced. You know, six, eight hours, maybe eight hours on the boat sailing there. We had to stand at the helm the whole time and hold the wheel. Right, hard work. What? Where’s autopilot? It’s not working. You realize how much you missed it. I mean, I still got there, but

Justin True: and you still had to get licensed and have the authority to know how to drive a

Ron Bockstahler: vote. Exactly. Yeah, because humans

Justin True: are. You got to fast forward a long time for when humans aren’t at the helm of life,

Ron Bockstahler: right? Absolutely. Well, let’s bring this back a little to attorneys. I want to go back to: Is there a red, light, red light, green light? If we’re talking to a solo or a small firm, attorneys, because that’s who you meet with a lot. Here’s the things I want you to start doing. Here’s things you got to stay away from, at least in the beginning.

Justin True: Yeah. Well, green light is non-customer data, back of the house business operations, your marketing, some of your your finance and accounting, anything else that just isn’t client data, just start there. Red light is client data,

Ron Bockstahler: right?

Justin True: Don’t don’t even use your mind space thinking how do I use my client data with AI, you’ll get there. Yellow light is is research, is publicly available information and research. Start trying to supplement what you would have done anyways.

Ron Bockstahler: And you, when you’re asking your AI bot to go do some research, you can ask your app, but to give me the proof sources. Also,

Justin True: it typically typically does automatically at this point, what you can do take it a step further is remember they’re they’re it’s your intern. Hey, I need you to go research X topic in X state, and I need X laws. I need that in this specific type of format. I when you cite the references, I need the links. I need the page. I need the think of the the biography or bibliography citing reference structure from middle school, right? I want it in this exact structure, or else you’re going to get an F. Like whatever you want, ask the tool for

Ron Bockstahler: right,

Justin True: because the tool knows everything except for what you know, and so if you want it to come in a certain format, just say it. I love it.

Rebecca Bockstahler: Well, if we were to take your CMO experience as well as your AI knowledge, and just go to like the very basics of people trying to market themselves, and at a point where answers pop up before a website ever would like, how does a firm get noticed in a web search?

Justin True: Yeah, that’s that’s actually it’s ironic. I I come from marketing. I don’t really do too much marketing anymore. My time is consumed in this AI world, and I need to find the answer to that question because I get asked it all the time. But my focus with AI is not on like ranking inside of AI. It’s it’s more of a culture and a company and an infrastructure and a mindset and a operations structure, which I try to focus on more things that are like. Going to have measurable results inside the business, just because there’s infinite topics in AI. Well, that’s the thing; you can’t be an expert on everything, right?

Rebecca Bockstahler: Have to find your one thing to master. Yeah, that makes sense. This

Ron Bockstahler: is we obviously there’s so much to cover on the show. We need to have your back on the show. It’s been so many things to talk about. I want to hit. There’s like a couple things before we end, talk to the audience about writing in their voice and what they need to do to allow their bot to do that.

Justin True: Great, great question. So, back to our intern. Hey, intern, I need you to figure out how I write. What is my writing style? For starters, ask me a couple of questions, right? That I might not think to tell you the answer to unless you asked me to kind of get a gage of personality. Once it does that, say, okay, now I’m going to give you some samples of my writing. You can go pull emails, you can go pull poems, short stories, deposition papers, whatever samples you want the tool to analyze to be representative of your writing style, ask it to build you a writing style sample pack and a guide. Now here’s your blueprint, like genetic structure. It’s and these come in the form. This is this is like 102 stuff. These are this is using what we call markdown files. Markdown files are just sets of instructions that call the shots, put the guardrails on, and you can have the tools make these markdown files. What I just everything I just said is is that hey, interview me. Let’s make my writing style guide. Here’s some writing samples. Make the writing style Markdown file. It’ll spit you back a one-two-seven-page document that is the genetic code of how you write. Top of there sits instructions. It says every time I ask you to rewrite something in my tone and my style, reference this file. And in the top of that file says always make jokes. What’s going to happen? It’s going to. It’s just going to constantly say jokes, and then when you change it to say never make a joke, it’s never going to make a joke ever again.

Ron Bockstahler: I love it. All right. Well, this is um. We’re going to have to wrap up the show. I want you like give the audience like here is why you want to call you, and then let’s how do we reach you? We’re going to put it in the show notes, but let you talk about it. Give you like two minutes.

Justin True: Absolutely, yeah. So I help I help law firms, businesses figure out who in their who in their company is ready and how how much. I help them figure out what AI can do inside of the business and who should be touching it and how. And then the biggest one is is how to do it safely and with what tools? This process takes four to eight weeks, depending on the size of the firm. We go inside the firm, we figure out how it operates, because even two identical firms do two identical things will operate inside differently. Different people, different standard operating procedures, and it takes knowing how information moves and people work together to really write the plan. So I help businesses, law firms, businesses and law firms navigate this world through through a four to eight week process. And yeah, I also have a monthly show I host an hour long Zoom that’s an open community session where I really just dive into some of these specific how tos so that people can get their paws wet and actually like touch the tools and feel confident about what they’re doing. I love

Rebecca Bockstahler: that it’s four to eight weeks because that’s something that people can easily see instead of thinking they’re you know it’s a two year process.

Justin True: Oh, it’s a it’s a four to eight week discovery process to really build the game plan for the next six to 12 months, at that point we we the the full document is it’s a playbook. We hand it over with the idea, and this is all discussed on the front end. The idea is that we’re going to walk you through this world. But if for any reason that changes, this document is your it’s your go. It’s your plan. It’s the same plan we’re gonna follow. So it’s it’s built very very detailed, very meticulous.

Ron Bockstahler: All right. Well, thanks for joining the show, Beth. It’s been great having you on. Obviously, got these questions. We got a lot more questions, so we’ll have to get back to those questions, and so we’ll have to have you back on real soon.

Justin True: Sounds good.

Ron Bockstahler: All right, that’s all for today. A big thank you to Justin, Justin True, for joining us today and for turning AI from a source of anxiety into a setting the guardrails for solos who actually didn’t know or don’t know how to use it. His information will be in the show notes. Please reach out to him. If today’s conversation got you thinking about how AI fits into your practice. I hope you’ll take just one idea, Justin, from Justin and put it to work for you. Fuck, man, I got to get better at this shit. You will. All

Rebecca Bockstahler: right,

Justin True: or we just got to do it in the evening.

Ron Bockstahler: If you enjoyed today’s episode, please subscribe. Leave us a review. Share it with a colleague who could use the information. I got applause because my teleprompter is going a little slower than I would like it to. The 1958 lawyers brought to you by Amata for more than 30 years. Amata has helped. Now it’s going way too fast. Has helped firms of all sizes do more with less. Build on our five pillars: fractional support staff, virtual office options, office space, legal community, and the marketing support. Whether you’re a solo practitioner just starting out or an established firm looking to right size your overhead, Amata gives you the infrastructure of a big firm without the big firm price tag. Learn more at AmataCorp.com. Until next time, work on your firm, not just in it.

Voiceover: Thanks for listening to the 1958 Lawyer. If you like the show, tell a friend, and please subscribe, rate, and review us on Apple, Spotify, or wherever you get your podcasts. If you’d like to hear more about Ron, Rebecca, or Amata, go to amadaoffices.com. That’s amattaoffices.com. All the links are also in the show notes.