Legal Tech Doesn’t Need More “Lamborghinis” ft. Kara Peterson and Richard DiBona

Speaker A: It's like AI can either make the justice gap smaller or bigger, and both could happen. And there's that tension right now. So I think that's where it starts to get really interesting.

Speaker B: There's also the part about leveling the playing field. We feel like what we're doing gives some of the powerful tools to people who aren't necessarily at, uh, big law firms. It seems like a lot of the AI is focused towards the biggest top 100 law firms and everybody's going after that big market. And we really want to make things more affordable and accessible to the other side of the market too.

Speaker C: Welcome to between the Briefs, a, uh, podcast by Steno. Um, we're here to bring you practical tips, expert insights, and real conversations about the pre trial process, court reporting, and the legal technology shaping the future of litigation. I'm your host, Adrian SEO.

Speaker D: And I'm your host, Joe Stevens. Whether you're an attorney, paralegal, or just curious about how technology is changing the legal industry, we've got something for you. Each episode will break down complex topics, share behind the scenes intel, and talk to the people leading innovation in and out of the courtroom. So grab a coffee and let's get into what's happening between the briefs. Welcome to between the Briefs your go to podcast for legal Innovation. I'm Joe Stevens.

Speaker C: And I'm Adrian Seah. Uh, today we're joined by Kara Peterson and Richard De Bono, co founders of dscryb, an AI company focused on making the law easier to find, understand and verify.

Speaker D: Kara is also the co host of Building AI Boston, one of the top AI focused tech podcasts exploring how AI is shaping the real world. Through conversations with founders, researchers and policymakers.

Speaker C: At uh, dscryd, Kara is helping build more transparent legal AI systems grounded in verifiable legal sources. Most recently collaborating with Anthropic on the launch of Claude for the legal industry. Kara, welcome to the show along with Richard. It's great to see you here. How are you doing today?

Speaker A: Great, thank you so much. We're super excited to be here. Thanks for having us.

Speaker B: Yeah. Hey, thank you.

Speaker C: Of course. Really great to have you guys here. So Kara, how about you give us a little bit of background. Where did it all begin for you and how did it end up, uh, being where you're at now? Building describe.

Speaker A: Yeah, so we have a really unusual founder story, although I suppose everybody thinks their founding story is unusual. It was the deep, dark days of the pandemic. Remember that? And we actually. So we're married co founders. So It'll make more sense if you know that from the start. So we were facing our own legal issue, which, you know, everybody faces at some point in their life, but you never expect it.

Speaker E: Right.

Speaker A: And so when it happens, you're like, what? So this was a, uh, discrimination sort of situation for where Rich was working at the time. And, and it was very startling and weird and hard to deal with. Right. Non, um, lawyers trying to figure out how to access the law. So just through what we saw ourselves, we decided that this really sucked and it needed to be fixed. So that's what we've been doing for the past three years. And there's obviously a little bit more to it than that, but that's kind of where we started and why we have this focus on access to the law and fairness and really trying to create tools that serve a lot more people than are currently served by legal tech.

Speaker D: Car this is an issue that's so near and dear to my heart. I'm a lawyer and I also teach at a law school and I run a public defender office inside a law school. It's the only one of its kind. And access to justice side of AI is something that we speak a lot about or we hear a lot spoken about and then sort of wonder where the action is. But you are actually sort of doing this. Can you talk. Give our viewers, uh, sort of a, a quick run through of what DSCRYB does and I'd love for you to also touch on the recent happenings between DSCRYB and Anthropic.

Speaker E: Yeah.

Speaker A: Okay. So I could probably talk about access to justice and AI for like hours on end. So I'll try to be as brief as possible, but I think it's giving probably the best opportunity that this problem has had in a really, really long time to move the needle. Right. And tend to make a big difference. To use AI to solve some of the problems that we have, which we know is a, uh, in a lot of ways is a supply and demand problem. Right? There's just way more supply, or excuse me, that's wrong. There's way more demand than there is supply for legal needs. We know that's huge, but we can't human our way out of that problem. Right? Because it's just too labor intensive, there's too much need. The cost structures don't work, people can't afford lawyers, et cetera, et cetera. So AI with the way that it can be applied to legal information and legal advice for some companies do that is the thing that could really change this forever. But what scares me a lot is that we won't ever find out how good it can be if we use it improperly now and we kind of mess up our opportunity because we're going to get regulations and things coming down that are going to prevent adoption. And I think we're already starting to see that a little bit with the dreaded, uh, hallucination issue. So we have, on one hand, this incredible opportunity to make such a meaningful impact on this very serious problem. On the other hand, there's so much risk and also aversion, risk aversion in the legal space that it's a really interesting tension. And I quite honestly don't know where we're going to land once we get through it. Uh, I think the jury's out.

Speaker B: There's also the part about leveling the playing field. We feel like what we're doing gives some of the powerful tools to people who aren't necessarily big law firms. Like, it seems like a lot of the AI is focused towards the biggest, what, top 100 law firms, and everybody's going after that big market. And we really want to make things more affordable and accessible to the other side of the market too.

Speaker A: Yeah, it's like AI, AI could either make the justice gap smaller or bigger, and both are. Both could happen. And there's that tension right now. So I think that's where it starts to get really interesting.

Speaker C: So, Rich, I would love to hear from you. How, how did you go about creating this? It sounds like it was. It started off as a, uh, epiphany idea and then, uh, through struggles, of course. But how complex was it to actually build it and get down into the nitty gritty? Because I understand you have a background in software engineering.

Speaker B: Yeah, yeah, my background is like, in big data management. So I was, I guess I didn't realize at the time how, how well that would be, how well it would apply. But the origin, as Kara said, was this legal issue we were facing personally. And so for the first, these, these legal issues, as Joe knows, could take four or five years. And so for the first two years we were, I was just kind of using Google and reading opinions on these books I would get from Amazon, like during the pandemic, sitting in bed reading opinion books and Cara's like, what are you doing? And so when ChatGPT came out at the end of 2020, 22, you start reading about it every day, and I was like, oh, how could I use this? Because I, I like to use the new tools as they come out and see if I could apply them to anything. But instead of just building a. A chatbot, which people were just building chatbots, I decided to take some of these opinions. I had been reading the judicial opinions and put them through this API that could summarize things. And so. And then I put, like, our court complaint through and summarized it. And I was like, wow, this is actually pretty cool. So then we're right around the corner from Harvard here outside Boston, and they had this. That case Law project. Is that what it was called, Cara?

Speaker A: It was. Wasn't it like, free case or. No, it was called the Case Law Access Project.

Speaker B: Yeah, yeah, yeah. And so they scanned in all their books. They had like 40,000 volumes, and they were putting them online for people because they were into free the law. And so I got access to their API and just started reading in opinions from Massachusetts and summarizing them. And then I made them searchable. And I think just a couple months in, I had like a couple thousand opinions and I had this basic search. And I'm like, carl, check out this cool thing I built. And do you remember that, Cara?

Speaker A: I do, yeah. Because I have a background in academic public health. I'm in marketing, but I did marketing for public health schools in town, BU and Harvard in particular. And I saw, like, access to justice is really like a public.

Speaker E: Or.

Speaker A: Uh. The lack of access to justice is really kind of like a public health issue, honestly. And so I just started getting really excited, like, this could help a lot of people and change a lot of outcomes. Not our tool specifically, but our tool plus, you know, other tools. And I think that's one thing we've really Learned over these three years, which is so long.

Speaker E: Right.

Speaker A: In AI years, that's like 500 years, is that, like, the more we work together as an ecosystem, the more likelihood we have a chance of succeeding. And, like, that's one of the best parts about this whole experience is meeting the other people in the legal tech area. And, like, the justice Tech group are, like, the best group of people on this planet. They're just amazing. So, yeah. So I quickly saw the potential, and I'm like, okay, we gotta make this thing out for the world. And that's where it started. And, you know, we certainly have done this. I guess I could say, Carl, let

Speaker D: me ask this, because again, I'm trying to sort of tie two things together. One is what you just mentioned explicitly up front is this dreaded hallucination problem.

Speaker A: Yeah, exactly.

Speaker D: And then we're obviously talking about this in terms of providing these, this information to people who need it the most and turning it into an access to justice builder rather than emphasizing the gap itself. But there is a truth too, right? I mean if, if lawyers who are supposed to be doing this correct correctly and knowing the law and knowing how to interpret cases are still all the time getting sanctioned or having to go to show calls, hearings because they are not verifying, they're not checking or they're doing things that wind them in front of a judge to have it explain themselves. How does that work though when you're giving this technology to people who might not even have the ability to actually verify the correctness of AI and to see if AI did give them some sort of opinion that lines up with the legal answer that is most beneficial to them. Talk me through that sort of tension.

Speaker A: Yeah, that's a really important point. And we are specifically in the space and we, we figured this out pretty quickly. You know, first we started, we thought we could work for, with like pro se people and you know, things of that nature. And we're very excited about that. But as we learned more about the ecosystem, we realized that wasn't actually the best place to make an impact, at least for the way we were building. We realized the better we could serve people, serving the people who need to be served is where we could actually having the most impact. So our tools, although people representing themselves can use them, you know, we're not going to stop them. It's really meant for attorneys. It started out, you know, as I say, being more direct to consumer. And there are people who are building wonderful tools in that space. But we are very much oriented towards helping the people who are helping the people. And so our thesis is if we can make your work, you know, you say you have a clinic like this, like your work more efficient, you can do more at once. I mean, tell me anybody who's more under resourced than legal aid providers, right? I mean I don't think you can find anybody, right. That's you know, trying to sit on the three legged stool, right. Or you know, it's is a three legged chair, I guess I should say.

Speaker B: Yeah, all stools are three legged, I think.

Speaker A: Yeah, I meant to say three legged chair. Yeah, you could, you could. Yeah, whatever. So. But you know what I mean, I had a broken chair. So really that, that felt like the place to build. So. But your tension is right because it is causing problems and solving problems at the same time. The question is how which is going to outpace the other. So Think about pro se litigants when they're putting in, you know, their own briefs like now they can sound very well written, right? I mean now in some way you can't tell maybe some, some of these things. So also the way that the chatgpt or clutter things will tell you that your ideas are good, right? It'll keep bringing you down these paths of like, yeah, you should sue. Yeah, you should file this. Yeah, you should. You know, this is right.

Speaker E: You're right.

Speaker A: What a great argument. Yes, the aliens were coming to abduct you and yes, you should, you know, put a brief together about that. And I'm not trying to make of, of the people's legitimate claims, but it does pose the risk of adding more junk to the system, which is already overwhelmed. So it's a great tool, but it has a downside. So in terms of people using it directly for their own cases, I, I, even though I'm super pro AI, I still worry a little bit about that, that if there's not that human, you know, in the lead, not even in the loop, like that that's risky now that's for like litigation matters. Now if you're dealing with a tenant landlord issue or you need like, you know, a protective order or there's some like, you know, kind of consumer issue, like there's things I think in the legal space that could absolutely be mitigated and handled with more of an AI thing. But I do worry about the idea of people just on their own. I really do. And it's kind of similar to health, right? Like yeah, in the old days you could go to WebMD and just sort of, you know, before I even like figure out what your disease was and go treat it. But that's a really bad idea. So.

Speaker C: Yeah, it does.

Speaker A: But I don't know, it's a tension, it's, it's a tough one.

Speaker C: It's never an easy to answer that for sure. And I would love to hear your perspective when it comes to the complexities of law. You mentioned you were building this out. Did you end up learning anything about how complex? Like it's the same like a simple Boolean of if then is not going to solve this. It sounds like it's, it's much more complex than that. Is there any lessons that ah, that came from when you were building?

Speaker B: Yeah, actually I hadn't thought about this until you just mentioned it. But so skipping back to where I built the first couple months and loaded things up and then I loaded in like all of Massachusetts or something. And then we decided to go and show it to our lawyer, or he had somebody in there and just like, look at this cool tool we bought. And we were expecting. He's like, oh, can it shepherdize? And we're like, what's Shepardizing? And he's like, it's useless without that. And we're like, oh, no. So we went back in and built, like, our own good law, bad law thing. And we. I learned a ton about that. And, you know, it's not like I'm doing all this myself. The disclaimer should be that what we were smart enough to do at the beginning was get all these legal experts and law librarians and everything as advisors. So every step of the way, we run things by people. But our naivety about how everything works has served us well. And trying to do things maybe an easier or better way, if that makes sense.

Speaker A: Approaching it, like, as a data problem, like, has given us, like, exposure to different ways to solve the problem. But of course, with. With the law, it's so intricate. You have to get it right. And the nuances are so important. Right. That that's where these people have come in. And they're amazing because, uh, I think legal librarians don't get the credit they're due. Although I think maybe now that's starting to change. Like, they're technologists, they're innovators. Like, they're doing incredible stuff. And so. And also they're really tough. So if you can get your, you know, legal librarian crew to. To approve the work you're doing or give it a thumbs up or kind of it passes muster with them, you know, you've built something pretty good. So that's been. We've always been his heart. Yeah, they give guidance, and, uh, they

Speaker B: give guidance on direction. Like, oh, this is nice, but you should also do this.

Speaker A: Yeah, but it is complicated. And then I think the other thing we've learned, our favorite thing is when someone posts that they vibe coded a solution in the weekend. Like, we always get a good laugh out of that.

Speaker B: Well, they could vibe code the front end the data. They can't.

Speaker A: No, I know. That's what I mean. That, like, oh, I. I just made Harvey in 24 hours or whatever. It's like, yeah, okay, sure you did.

Speaker C: You mentioned a very key word there. And, Joe, I think maybe you'd be the perfect person to elaborate on this, and that's shepherd eyes. From your experience, Joe, could you give us, like, the definition of what that may be for our audience? That, uh, we don't understand.

Speaker B: Oh, sorry about that, Adrian.

Speaker D: Yeah, it's basically using the database. Like, let's say you ask a legal question and then you get an answer. The question then becomes whether or not the answer you just got is good or it's been qualified or it's changed or, you know, a, uh, higher court said that no, you know, you're partially right, but partially wrong. So you need to know if the case that gave you the thing that sounds good still is going to be compelling or binding. Uh, so. So, yeah, that functionality matters tremendously. I can't imagine the complexity of that.

Speaker A: Yeah. And Rich even took it a step further. Right. Rich, with the, uh. So most of these citators. Because Shepardizing is like trademarked word. Right. So most of these citators are doing it at a case level, but Rich realized with all the data we have, he could actually do it at an issue level. Rich, tell me about that, because that's pretty fun innovation.

Speaker B: What we did is we went through every single place that a case cited another case, and we figured out what kind of treatment that was using AI. So basically, if you think of the human editors that have been doing it for a hundred years, we just had a bunch of AI editors do a 24 by 7 for six months and run up a huge OpenAI bill. But we won't get into that.

Speaker A: But, uh, that thing floating above Rich's shoulder, there is plaque from OpenAI. How many tokens was it?

Speaker B: It was like 100 billion a year ago. I think we're like.

Speaker A: I'm like, Sam should come over and make us dinner.

Speaker B: Yeah, I got like a YouTuber plaque right there. That's amazing. Yeah. So we compiled all these treatments, but because the AI isn't getting tired and it doesn't care, we decided instead of just saying that this whole case is good or bad based on this site, tell us what issues that it actually overruled or followed. And so what we found is that there's a lot of cases that might be overruled on one issue, but the same case follows it on something else or says, you know, we're not going to follow it on this, or we just mention it on this. So, you know, I'm not sure if it would reopen any cases for use, but it certainly is informative. Like this is the issue that they're actually overruling. It's not just the whole 40 pages was overruled.

Speaker D: Yeah, I can immediately think of the use of that. And it ties into a question that I have And I suppose, I mean, I think I know how you're going to answer this, but I would just. I'm curious. You know, obviously it seems like, you know, when the foundation models get good enough, everyone's worried about, okay, is that just going to eat the layer that has been built? Right? So the question becomes like, what's the moat there? It sounds like the moat is the data, but is that true in your case? And then I think it touches on, um, connected to sort of the first question I was really asking as well, which is all about, you know, how things have evolved with anthropic and what that partnership, the evolution of that and where things stand now.

Speaker B: I was actually thinking about this two nights ago, so there's kind of like I could answer both at once, because what I realized is there's like four levels within Claude now of how accurate could get. So the level one is like you just open up Claude and you ask it a question, and it uses whatever is in its model memory, so whatever it was trained on. And so it may or may not have statutes, it may or may not have regulations from certain things. It definitely doesn't have all the treatments and everything that we were just talking about. Level two is Claude plus the Internet. So you let Claude go out and it searches some websites, it might search Reddit, it might search somebody's blog post about this. And so you get some. A little bit more accuracy because it might include some Google Scholar, but it's still not the best you could get. The third level would be, like, where you give it raw opinion data. You're like, here's judicial opinion PDF, give me some analysis of this. And it does it on the fly. And then I like to think of ours as the fourth level. And this would be if. Also if any of the big players expose their data to cloud, which I don't think they do, but this is what you might get in co Counselor Vincent or something like that is where we took what we ended up with after slicing and dicing and analyzing. We have like 300 million data points in our database now. And so, um, when we give Claude some tools and we say here's tools you can use, and Claude says in response to somebody's question, it'll say, oh, I want to know if this opinion actually says this quote. And so we say instantly, yes, this says this quote. And it also says this and that. So we return data right under Claude's nose that it can use. And those are the best possible answers because it doesn't have to Go get them off the Internet doesn't have to try to read through a 40 page opinion on the fly and miss some stuff because uh, you know those opinions are pretty dense as you, as you know, Joe and you Adrian. So just getting the data as close as, and as close to the user question and as accurate and curated as possible, if that makes sense.

Speaker D: It does.

Speaker A: And then like is our uh, you know, OpenAI and Anthropic and, and, and other players, Google and whoever else comes along going to eat everything up? I mean that's the biggest question of all.

Speaker E: Right.

Speaker A: And I think the answer is they're going to eat a lot of things up. I don't think they're going to necessarily eat everything. I mean they may be able to, but I don't know. That's this weird dystopian future where there's like two or three companies that do everything in the world. Seems like kind of like a horrible place to live. So we'll see. But in some ways it's their decision where they stop.

Speaker E: Right.

Speaker A: And where they want to stop. And so we'll see. Right. But it's a lot of work to build this stuff and just from a business perspective, like I don't know if they'd want to, but we'd have to ask them if that's what they want to do. But you know, they'll be able to eat a lot of things but I don't think at least right now they could eat up people like us or necessarily Thompson, Reuters or you know, some other purpose built legal tools. And I don't even know if it would be in their benefit to do so honestly, because they can't be experts in all things. Right. You're just, you're going to lose nuance if you don't have people who know how to build these things and you kind of need to build a community. And so that's where I think Anthropic made a really smart decision in what they are doing. The way they approach the legal, the cloud for the legal industry launch. So happy to.

Speaker B: And other verticals.

Speaker A: Yeah, and other verticals. But yeah, but I specifically know about the legal one and I think what they did to have a special curated focus, public service slash access to justice tools was brilliant and um, I'm very pleased to be part of it. So happy to tell you more about how that came about.

Speaker C: Yeah, I do want to know, learn more about that. So how did this work out? There was a representative of Claude going hey, we see your database. You know, this is exactly what we need to anchor our responses to, not hallucinate. What was that approach like when it came to trying to collaborate with your company?

Speaker A: Yeah, so I would say there's. Okay, so talk to. Talking about moats again. So I see like two moats right now for companies. Right. So there's the data one and not just data, but structured, cleaned, really like ready to use out of the box data. Right. So there's that. But the other moat is sort of like reputation and trust. And I think that in the legal tech industry, like that's really, really important. So our foot in the door with Anthropic actually wasn't through the data side, although that had to pass muster obviously with them once they do about us. But we actually got in through our reputation and trust. So what I mean by that is we have friends out there in Everlow for good and the Justice Technology association and the Free Law project and courtroom 5. Like so other people in the community who, you know, advocated and said you need to see these, these people and what they're building and they'd be perfect fit. So honestly, they came to us with, with the, with the ask to collaborate and it was obviously a very great day when that happened. It was very exciting. So that made us feel really good because one of the things we prioritize is building with our community being super transparent, like fair pricing, uh, really being part of the ecosystem. And it felt good that that came back to us in a good way. But then of course, then this stuff is like, okay, we want to look at what you have and then, you know, then you have to be good enough to be part of their, their, their ecosystem. And some stuff that Rich can tell you about since then and how well it's been being utilized by Anthropic itself has been really exciting. Rich, I don't know. Do you want to talk about the, the GitHub stuff?

Speaker B: Yeah, I mean, I don't want to step on another one of your questions first or. Oh yeah, I could give you one, please.

Speaker D: Yeah, go for it.

Speaker B: Three sentences or something. So they, and this was all a surprise to us. Like they didn't come to us about this, but so they have their whole anthropic GitHub repository with all their skills. And this was the thing that like two months ago everybody was in a big, you know, it was all the commotion within the industry. Yeah. Where they came out with these legal skills skills and it caused all people's stocks to go down and they were like, oh my gosh, Anthropic's gonna enter verticals. And so on launch day of this Claude for legal, I actually saw that they mentioned describe 64 times in their thing as like, if you're gonna do this skill for step one, download dscryb or, you know, they mentioned a couple. And so we were mentioned in all these skills and it's like, oh my gosh, like, it's distribution.

Speaker A: Like go, uh, upgrade the servers.

Speaker B: Yeah. From a business perspective, like, we had everything going except awareness because we're so small and we couldn't, you know, we can't afford to just blanket the country with ads and everything. So it really took care of this distribution. We went from this cool indie band that people like that they go and see and to now we're like you two or something. Not really to that level, but you're a lot.

Speaker A: You're a lot taller than Boho. Uh, yeah.

Speaker C: Hey, you know, that's incredible. I.

Speaker D: It's what, what an amazing story. And to hear this evolution, I am just so interested again, the lawyer in me. You get on LinkedIn, you whatever. Somewhere, some right now, some lawyer is out there not doing what they should be doing, and they are putting something in a brief when we all know that every state bar, every judge that has had any opinion, like, we're talking about the need to verify your output. You talk about so easy. Yeah, but so, so how do you. Are there things that you do or, uh, how do you think about this? Are there ways that you try and emphasize the need to do that or like, how do you participate in this real disaster zone?

Speaker A: Yeah, well, my dream is to like, get the court. And it's so hard. I don't need to tell you this is. But to get the courts to kind of understand how it's in their interest to actually use some of these tools. And you know, some states are much more progressive than others and, and all that. But one of the things we did early on is we actually were like, all right, let's create a tool for that. We'll build a tool for that. So we have the car.

Speaker B: To your point about the.

Speaker A: Yeah.

Speaker B: The courts being progressive or not, it's not a political progressive because some of the most progressive states.

Speaker A: No, I don't mean progressive that way. Yeah, uh, yeah, I mean, like, innovative is probably a better word then. So. And this goes back to the worry that I was saying before, that if, if judges or, you know, people that, uh, can make the decisions about how this stuff is going to be used start to see it's just creating a giant mess. You know, judges aren' necessarily the most technically savvy people. Like, we know this, some are, some aren't. But courts don't have like, you know, giant IT departments to innovate and do all this stuff. So it's concerning that. It's like if they only see the bad side of it, we're not going to get to where we want to be. So. But checking for hallucinations is so easy, right? If you have the right tool. So we decided we're like, let's create a tool. So on our platform, there's m more bells and whistles on the platform than in the Claude connector. You know, there's some parts that are pretty close. But like one of the tools we have on our Plat platform is you put a brief in and it will tell you either yours or someone else's. Is everything in there good? You know what I mean? Like, Rich can explain it more. So basically it's like there's really no reason to not use the tools that are out there. Just make sure you're using good tools.

Speaker E: Right.

Speaker A: Otherwise it won't help you. So Rich, maybe give like 2 second overview of the brief checker and then tell them about challenge, because that's fun too.

Speaker B: Yeah, I mean, we have a deep research tool where this is the difference between our platform and our cloud thing. And I don't want to go infomercial, but it's actually kind of, it's actually kind of cool that the brief checker, what it does is it not only checks if the citation exists or not, but it also checks if you're actually citing the source material, like quoting it correctly, if you're paraphrasing it in a way that is different. Because what we've noticed from testing, like thousands of briefs through this thing is that it seems like attorneys like to paraphrase things from the original instead of direct quotes. And sometimes they may like, you know, move it over just a smidge to suit their needs. And so this thing will call that out and maybe it's helpful before you submit it as your own or if you get the opposing one. So it tells you if the source material supports it and it tells you what the original, uh, opinion you cited is usually used for. So it's pretty cool.

Speaker C: I want to hear from you, Kara. Uh, when it came to building this out, were you guys prepared for the level of, of like the volume of users that would just start using this all of a sudden? Because the use case is obviously phenomenal, right? And the average person now can just look into this, look up their, their case, compare it against various, what's out there, existing and uh, see if it works. Now were you prepared for this level of interest, especially with the collaboration of Claude and how did you go about dealing with that as a business owner all of a sudden?

Speaker A: Yeah, that's a great question. And yes, I mean, yes for sure. But like the growth. So it's been a month, it'll be a month on 12 June 12 since launch. So it's, it's still really new but like the growth has been really, really crazy and, but it's great. But I can't give away the secrets because Rachel got mad at me. But the way that we built our tool, the more people use it doesn't necessarily matter to us. Like there's, there's scalability in there that doesn't increase our costs or like do things. So the way it's built, unlike other AI tools, it's not like, oh, you use your AI allotment for the day like it does on um, 1M, part of the platform. But like when you're in cloud, it doesn't do that. So in some ways scale doesn't matter. So the only place it really would be, you know, difficult is just the more questions we get. And we're very dedicated to like we say we answer all the questions ourselves. Like we talk to people directly, like support questions. Support questions. No, no, no, my name is Claude. No, no, I do not, I'm not pretending to be Claude. No, I mean support questions. So you know, sure. We're at a point where, you know, a two person plus advisors, bootstrap, self funded company. Like we obviously, you know, we're in the very last point of that being who we are. So we're enjoying it at the moment. Right. Meaning we know there's a lot of demand. So we're going to be growing and that's great, that's exciting. But it's also, you know, it introduce risk and all the other things, but yeah, but it's a fun problem to have. I'll put it that way.

Speaker C: Definitely.

Speaker A: Yeah.

Speaker D: On the leadership, the building side, the bootstrap side, uh, you know, you've got your advisor. It sounds like you've really done this thoughtfully. Uh, did you have any sort of guiding principles, either of you, that really, you sort of were like, as we go down this uncertain road, we want to make sure that we adhere to some fundamental truths or some fundamental principles about what we're trying to create yeah,

Speaker A: that's a great question. And absolutely. And this is where being married to your co founder, as long as you have a good marriage, I suppose, is, is a, uh, is a nice thing. I mean, obviously there's downsides. You never stop talking about work and things like that, but you pretty much are aligned on your value system, right? So it's like, what are you trying to do and why? And so there was a time where we were like, do we want to be a nonprofit or do we want to be a for profit?

Speaker E: Right.

Speaker A: And we actually decided, okay, we're going to be a for profit or a no profit or a no profit. Right. That is probably the last, that would be the last option. But you know, we want to build something that really matters. And not only matters because it's a viable, you know, business and we, and we do well, but that it actually has an impact out there, even if it's small. We want to do something to make things better. And so that is always our guiding star. And we also really, really believe that you shouldn't have to pay huge amounts of money for these tools. Uh, the way legal tech is priced is really weird when you're not from that market and you look at it and you say, this is absurd. Like, we're $50 a month for the flat platform, $25 a month for Claude. That's normal for other kinds of software, right? Like, it's not weird for other, uh, kinds of software. It's, it's, you know, Canva or like Descript or whatever you're using. Like that's normal. Legal tech is really weird. And I get it. It's because there's a lot of, like, human work and whatever, but it doesn't have to be that way. Right? So we are, aside from our sort of mission, our product promise or our thoughts are you don't have to trade really good tech for fair pricing. Like, that's just. We don't. You don't have to anymore. You don't have to. You don't have to. And we also believe in not over engineering, probably because we don't have the bandwidth to, but also because not everyone needs these gigantic end to, uh, end solutions. Now. Some places do. If you're a giant law firm with all this really complex litigation, yeah, great, go for it. You can afford it. Have at it. Like, there's a lot of great options, but small solos, you know, nonprofit, whatever, they don't need all that. And it's, it's not fair to the market or it's not smart even as a business to be like, we're only going to build something that's like, I don't know, like a Lamborghini. You don't need a Lamborghini to go down and get like a gallon, um, of soy milk at the corner store.

Speaker E: Right?

Speaker A: Like you don't. You can just ride your stupid bike or whatever.

Speaker B: We have more than a bike.

Speaker A: Why not?

Speaker B: We have more like a. Probably like a nice little Lexus SUV or something. Yeah, I can.

Speaker A: Feeling a little facade or something. Do those, do they even make those anymore? I don't know. That's my, that's my Gen Xer in me.

Speaker C: When it came to building this, is there any advice you would give to any other, I guess any of the other aspiring, you know, AI developers or developers that want to build on top of AI, Anything like you learned or advice you would give to otherwise, even outside of the legal field?

Speaker B: Yeah, I mean, it's really hard. I mean, they say, I mean, you read about how hard it is and it's every bit as hard. Especially like, as Kara said, we didn't take VC money checks or whatever because we always wanted to. We've had them knock on the door and say, oh, this and that, but we've always. We don't want them to tell us to triple our price and we don't want them to control what we're doing and say, oh, don't bother serving that low thing when you could charge 250amonth for your thing. That's 25, you know, so we've always been very careful about that. So it is hard. I'm, um, in the age of AI, I think. I mean, I'm lucky that I have like a software engineering background. So I, it's. Being a programmer now is more like being a project manager. So, you know, I don't even know what kind of advice I would give to somebody who doesn't know how to program and is just starting off because, uh, it's actually really hard. I mean, it's hard.

Speaker A: It finds.

Speaker B: Yes. Find somebody who does.

Speaker A: Yeah. Like partnering up is so important like that to me. Like, the founders that have the best shot are the ones who have found the right partnership. And, uh, you don't have to be married. You're, you know, co founder. That's. There's other ways to do it. But like, you know, find your missing piece and if you don't know it, like, figure it out. Right. Because if you have these like great companies that start with a bunch of really smart engineers that's great. But who's doing your storytelling? Right? Who's doing your marketing? And storytelling right now is huge. You know, your podcasters. We gotta do our own storytelling. And if you just have great tech, it's really hard to get noticed. But don't be dissuaded. Like, look for the big problems, the problems that seem almost unfixable, and start there, because that's where some cool stuff's gonna happen.

Speaker B: And don't cut corners. I thought it went well.

Speaker A: Don't cut corners.

Speaker B: One thing we never did was we never sold, like, unlicensed data or commerce. You know, there's data sources like Cornell, and we see people trying to sell that commercially, and. And we don't like violating terms of service. That always comes out at the end if you do that type of stuff. So we never cut corners. We made sure we did everything the right way. I would say that's good advice. If you want to build a good company for the long term, you got to do it the right way, I think.

Speaker D: I can't tell you how much I love what you just said. I mean, in a world where it's so easy to just tie. Make, uh, a leap from AI to return, just some sort of financial return, and we're talking about profitability or we're talking about scalability or whatever it might be. I love. I think what you said earlier, Kara, resonates even more deeply because I do think that that AI for good in the legal space community feels different. Uh, we had, uh, you. Maybe you know him. We had Satish Nouri on here.

Speaker A: Oh, yes, he's great.

Speaker D: He's wonderful. And there's just an energy. There's a recognition of the problem. People are trying to tackle it. It seems so outsized. I think it leads to maybe my last question to you both, which is the reference point. Uh, people talking about AI being so disruptive to the legal sector. And then usually when people reflect on it, they say, you know, maybe the most big. The biggest revolution prior to this was the digitization of case law. And then they talk about Wet's Law and they talk about Lexus. So you talk about these two giants that really have had sort of a stranglehold on this for so long. And, uh, now things are kind of changing. Do you get a sense of how you're perceived by the more deeply rooted institutional players?

Speaker A: Yeah, surprisingly. At least, how it seems is that they like us.

Speaker E: Right.

Speaker A: And I think it's because we didn't make the mistake that I think some companies do and go out and say they suck and we're gonna, you know, take them over and we're gonna be. It's like, no, they, they don't suck. Stop it. You know, and you're, and no, you're not gonna become that company tomorrow. Like, cut it out. And I just actually had the chief product officer from Thompson writers on the show and we talked about some of this stuff. So. Yeah, and it was really, really interesting because, you know, obviously we were joking around. Like we're in the same pool, we're just at different ends. So I think it's, you have to be a little bit humble when you're a builder and like, you need to recognize that this is an ecosystem. There's a place for Thompson writers. Absolutely. There's a place for Lexus. There's a place for, you know, Clio for sure. And I love Clio and I love the Velux team and the Fastcase team. Like, those are great people. They've always been friends of ours. So it's just, just, it's. We can solve these problems together, right? I think better than alone. And so we have to think about it that way. And I mean, everybody individually cares about this stuff. It's when this whole idea that the only value that you can create out of an AI company is to be the richest person you possibly can, like, I'm going to call bullshit on that and I don't know if I'm allowed to say that word, but I just did. You can beat me if you need to, but we don't have to assume that the only good outcome that you can have as an AI founder is being like the next billionaire. There's other things you can do. You can be really, really rich and make a lot of good things, but you can also do something really cool for the world. Like, and you don't have to either be like this impoverished, you know, like non, uh, profit person on the four legged, three legged chair or like, you know, Elon Musk's, you know, ask person. Like there's a whole lot of room between. So I don't know. Making more tech bro billionaires doesn't have to be the outcome of legal, the legal AI revolution. Like, it would be really sad if that's all it was. And that's where about 99% of the energy and the hype and the glazing on LinkedIn is going to is. Who's the next, you know, tech pro billionaire out of legal tech? And it's just nauseating, to be honest.

Speaker C: Well, I Think that kind of brings us to our natural next question. And we always ask this at the end of the conversation, and that is. Is what is your hottest take on the legal industry right now?

Speaker A: There you go. So let's decide what really outcomes do we want? Like a few rich people, like, no. Unless you're one of those few rich people, you know, which you're not going to be. Like, let's figure out other outcomes. So there's that. But I also think the kind of most interesting thing playing out right now is where people are going to be trying to serve the attorneys and the paralegals and the people working in this space. There's the idea of that. I've, um, got the walled garden. You got to come to me, right? And that. That's ex. Those are the really expensive tools. Or it could be like, hey, you're over here. I'm going to come to you. Right? So. And some companies are doing both. We, as everybody knows, like, because we just talked about. We jumped all in and we're like, you're in, Claude. Like, that's great. We're coming to you and we're going to make. You're using it anyway. Let's make it safer for you to use it, right? Let's make it better for you to use it. So that way, you know, that's good for anthropic because it's sticky for their people being there. It's good for us because people are using our. I think you gotta follow where people want to be. And I think the companies that are going to really make it long term are the ones who are going to understand that, that we are not going to be in a space where, you know, Ma Bell is the only one in town, which I will see how many people know on the call what that means. So that's my hot take.

Speaker B: I. We are so married. Because she completely stole my answer. I was going to say, like, from like a tech perspective, it seems like to me this cloud thing, like with these connectors and you could do similar things on some of the other tools. Now it almost seems like when they launched the app store for the iPhone and people stopped building their own things outside and they started building these things that work right in the phone. And we still see that to this day, although now we do go back to the web a lot. But for a while, the talk was, our, uh, website's gonna die and everything's gonna be on the iPhone. And so we kind of used our experience of living through that. It's one of the good things about being a little older founders is we actually remember 2008. But yeah, I mean I, I kind of drew that parallel in my mind and was like, we, we should get into that on the ground floor. And so far it has paid off.

Speaker A: Can I give one more hot take just in addition?

Speaker B: Yeah, you give my hot take for me.

Speaker A: No, no, I'm in. Your, your hot takes. All good. So I think the ethics of these AI companies are going to matter too, a lot. I think we're seeing that now. And you know, I think anthropic, you know, is gaining share in the market and doing really well because I think people do want to align with tools that they feel more ethically aligned with. Um, so I hope that's true. I feel like we're starting to see that. But you know, consumers will vote with their feet and they, and their clicks and they are. And, and so hopefully that will mean that using some of these tools to solve big problems like access to justice will be not only a good thing to do societal wise, but for their reputation, their bottom line. So hopefully that all can merge together in some problem solving.

Speaker D: I just love this. It's so wonderful to interview two people who are just so committed and have the ethics in mind and have the purpose in mind and, and it's just really refreshing. And uh, with that, that's going to be a wrap on this episode of between the Briefs and a big thank you to Cara and to Rich for joining us here today.

Speaker B: Thank you so much.

Speaker A: Thank you.

Speaker D: It's, uh, such a, such a joy. Yeah. Adrian, any final thoughts from you?

Speaker C: No, this has been an incredible conversation. I thoroughly enjoyed it. So once again, thank you, Rich. Thank you, Cara. This has been great. It's been really nice hearing about how we are improving our access to justice, uh, in the legal industry. So on that note, be sure to subscribe so you don't miss future episodes. Thanks for listening. Stay curious, stay inspired, and we'll see you next time up. Uh, next on, um, between the Briefs.

Speaker E: For me it is about self reflection, being self aware, understanding again, what my strengths are, where I have some things that I need to work on and really getting an understanding of who I am. And then also I have always, always talked about this idea of having a career or a leadership track that has a gps. I know my goals, I understand my purpose, and then I understand what skills I am great at and what skills I need to build. And that GPS is going to guide me to be the type of leader I want to be. So if I was giving advice to someone, I'd say, sit down with yourself. Understand who you are. What is it that you're trying to do? Let's put those down into some specific goals. Let's understand ourselves. What are my values? That's how I define my purpose. And then is there a skills gap or are there skills that people always compliment me on? Let's leverage those. And that is the roadmap I use to become the kind of leader that

Speaker A: I want to be.

Speaker C: Stay tuned for the full interview coming to you soon. Between the Briefs is brought to you by Stenna. To find out more about Steno and how we combine exceptional court reporting and litigation support services to deliver a superior litigation experience, visit steno.com that's s t e n o dot com and then make sure to search for between the Briefs in Apple Podcasts, Spotify, or anywhere else you get your podcasts and click subscribe so you don't miss any future episodes. On behalf of the team here at Steno, thanks for listening.