Rebuilding the Law Firm for the AI Era

Hey, everybody. Welcome back to the Dominate Law Podcast, where we empower attorneys as entrepreneurs. I'm your host, Donna Desha, and today we are tackling one of the biggest threats and arguably the biggest opportunity facing the legal profession right now, the rise of artificial intelligence and what it means for the future of your firm. Before we get into it, a quick word about our sponsor, echo Marketing. Here is the reality nobody talks about. AI is leveling the playing field. Inside law firms, big firms, small firms, solo practitioners, everyone now has access to tools that beat 99% of humans on the bar exam. What separates the firms that win is no longer just legal skill. It is visibility. It is being the first name your future clients find when they search online. Equa builds exactly that with proven digital systems that attract the right clients to firms that deliver real value.

They're giving every listener a complimentary one-on-one marketing strategy session worth $900, where a senior strategist researches your firm before the call. Book it at www.dominantlaw.com/msm. Now into today's conversation, we are calling this episode, rebuilding the Law Firm for an AI era across four pillars. We will explore what AI can genuinely do today versus where it falls short. The new skills junior lawyers need to compete, why the billable hour is collapsing in front of our eyes, and the ethical risks every firm needs to navigate. By the end, you will have a real blueprint for building a future ready practice to guide us. One of the most influential thinkers in legal tech today, Damien Rio, federal court clerk, 15 plus year litigator technology lawyer, and one of the architects behind Vincent, the AI legal research platform, built on a billion legal documents. At v Lex, he has built an ontology of more than 18,000 legal tags that is standardizing how the entire profession thinks about legal data. Even Elon Musk, before his political turn publicly called him out and said, this guy rocks. Let's dive in. Right? My first question to you, in today's legal practice, which tasks has AI proven genuinely reliable at handling, and which core legal functions still require human judgment?

I think that, uh, you know, a lot of what I went to law school for, and a lot of our listeners that went to law school for, um, are being largely done by AI today. Uh, that is, I spent a lot of time working, uh, you know, in law school, and, uh, they would say, you know, be sure that you go to the stacks of the books because, uh, they said back in 1999 when I was a first year law student, they said, you won't always have Le Westlaw and you won't always have Lexus. Uh, so you'll want to be able to go to the books just in case you don't have those things. Uh, so anyway, um, that, uh, that, uh, opinion didn't last long because I've, uh, always had electronic resources and me going to the stacks didn't really, uh, make me a better lawyer.

It just wasted my time. And so I think that what, um, what we're having now with AI is AI could be able to read through, uh, you know, uh, 12 hour legal research task that it used to take me 12 hours, and it literally does it in a minute and a half, uh, where, and, and it does not just as well as me, but way better than I can because when, uh, uh, I would maybe read when I was litigating a few cases and then find the ones that would work best and make it into a brief, but in the same time that I would read a few cases, uh, Vincent reads through hundreds of cases, lands on the 25 that are above a 70% threshold, and then is able to give a, uh, a really good 25 or 30 page brief, uh, that is on top of this in about a minute and a half.

That's less time than it would take for me to read half of one case. Uh, and so really, I, I think that, uh, when we think about what we as lawyers do, it's not doing research for research sake. We're actually doing research to be able to reach an end goal, to be able to win the case, or to be able to draft the contract. Um, and so if you think about, uh, you know, legal work not, uh, and research not as a means by itself, but a means to an end, I think that we, uh, as AI can actually give a lot of the end too <laugh>, you can actually make the documents. Uh, we, we provide complaints, answers, uh, motions, that kind of thing. So really what's left for us as humans, you know, if you think that your legal work was providing widgets, that is the motions, the briefs, the pleadings, or the contracts that you, uh, the widgets, um, I think we're gonna be sadly mistaken because, uh, the widgets AI is gonna be taking all the widgets.

Uh, but really what the clients are buying is not really a widget from you. They're buying trust. They're saying, uh, Hey, uh, I'm your lawyer. Your problem client is now my problem. Uh, that is, you can trust me that I'm gonna be able to shepherd you through these hard problems that you have. So we're selling trust, and I think that humanity really is what's gonna, uh, come to the fore going forward, is that, uh, you know, uh, do you trust chatt PT to give you a non hallucinated answer, or do you trust Gemini or do you trust any ai? Uh, open claw is, is the agent of the day, uh, and you probably trust them less than you trust the human, uh, because you can actually look that human across the eye. Uh, and there's a, my friend, um, who is a Jordan Furlong says that there's probably three aspects of humanity that are gonna be, uh, left after the AIS takeover.

Uh, one of them is trust, uh, that I, you can trust me to have the problems. Um, and number two is to be able to say that, uh, integrity to say that, uh, opposing counsel knows that I'm a person of integrity, and the judge knows that I'm a person of integrity in the way that they won't necessarily do that for, um, for ais. Uh, and then, uh, so trust, integrity, and then intuition is the last one. Uh, because you're not gonna be able to, if you're in front of a judge, you're not gonna be able to have chat GPT in your ear as you're arguing in front of a judge, and you're not gonna have chat GPT in your ear as you're around the negotiation table for that. You need intuition. Uh, and so I think that trust, integrity, and intuition probably are gonna be the three aspects of humanity that will persevere, even as the ais are doing a lot of the lower work, uh, that is making a lot of the widgets.

And, uh, the last thing I'll say in the last 30 seconds that we have is to be able to say, you know, we, um, are gonna go from ideation where we make a lot of, uh, we sit back and we think really hard. And in that same time that we think AI can and maybe have three ideas, AI can come up with a hundred ideas. And so what's left is we're not ideating. We're actually curating the good ideas with our good taste. And so I think we're gonna go from an ideation, uh, starting from a blank page to a curation and being able to give our taste.

Okay. I apologize for that. We are kind of testing a little thing over here, but, uh, Damien, I am curious about one thing you mentioned over here, right? Uh, with regards to how things were done in the past. Um, you know, you would go to the stacks first and spend a little bit more time looking at, uh, cases that might not be as relevant as you mentioned, like above that 75% threshold, right? What are we losing there? I mean, there should be something that we are losing there, right?

Maybe so, like, the reason, so what, what is the reason for researching? Uh, part of it is to understand the law, right? To get a sense of, you know, where the law, uh, says you can do something, or where the law says that you can't do something. And by reading cases, we kind of take the law in by osmosis <laugh>, where, where we think if we read enough cases, we will just absorb, uh, what the law is. And so I, I think the basis of your question is say, well, if we're not absorbing like that, are we really losing something? And I think the answer is maybe, uh, because we're kind of losing that intuition that if you read a thousand cases, then you'll get intuition as to what those thousand cases do, um, in a way that you can't, that if you prompt chat GBT or an AI model like mine a thousand times, uh, that doesn't really stick in your brain in the way that, uh, it does to wrestle with the cases like in the past. So I think that's maybe one thing that we will lose, is to be able to, um, lose our intuition. Um, and there's a real question as to how important our intuition is. I think it's really important. Uh, but I also thought it was really important to do use a calculator or to, to do mathematics in my head. Uh, but calculators can do that all the time. And maybe losing calculators is less important. Uh, but again, it, uh, I think that it intuition probably is more important than calculators.

Thank you very much for that clarification, Damien. Now, where do you see the biggest disconnect between what AI tools are marketed to do and what they can realistically deliver for lawyers?

I think, uh, you know, we are all going through maybe the most exponential growth of technology in our lifetimes. Uh, where, uh, the, you know, if you had said, let's say before chat, GPT came out in November of 2022, if you would say, in less than four years time, in three and a half years, uh, we are going to have AI beat 99% of humans on the bar exam. Uh, they would've said, you're crazy. Like, this is, this is, uh, that, that's, you're a fool's errand. This, that will never happen. But here we are, just about three years later, and it has, that has happened. Uh, and we went from horrible, uh, images like, you know, AI created images were just like, you know, I don't know if you remember Will Smith eating spaghetti, uh, but it was just like, it was abhorrent how awful it was back in 2023.

And now, um, they've done double blind studies to say that people can't tell the difference at all between the AI generated in the human. So anyway, so to your question about marketing, I I think that, um, you know, people that were marketing that AI was gonna beat 99% of humans if they were marketing that in 2023, people thought, oh, that's hype, but here we are. And it's true. Uh, and so I I think as we reach this exponential growth up, I think we're all gonna have to reassess like, what is it, uh, that is truly marketing hype? Or what is it that AI can actually do it? But I'm just too busy heads down doing my work, and I'm not looking up to realize what's happening in the world around me. Um, there's a, there's an evergreen cartoon, uh, where, uh, uh, someone is, uh, taking a cart and the cart has square wheels, and somebody comes to them with a round wheel saying, here you go.

And the square wheel person said, no, no, no, I don't have time. I'm too busy. Uh, and so I think that the, the AI is the, the round wheel that's being given to the square, uh, wheel person, uh, and the round wheel is now going exponentially faster, beating humans on the bar exam. And I, I think that, you know, AI tools, um, there's a, uh, Anna Gu, GUO is the name of a, a researcher, I think she has Law bench is the name of her study, but she does benchmarks to be able to say, humans do a thing, and then the machines do a thing, and then they mix them up and then give them to human reviewers. And she said that chat, PT out of the box beat humans, uh, in the humans. Uh, the human reviewers said that the, the machines, Gemini and chat PT were both more reliable and more useful than the humans.

And that is chat pt outta the box. And that is Gemini out of the box. Mm-hmm <affirmative>. So if you think about what, um, legal technology tools do on top of those, um, of course they can only do better, right? Because that's chat, chat. BT was not made for legal, and Gemini was not made for legal. So, but if those tools build upon chat BT and build upon Gemini as they do, um, how much better can they be than humans? So then we have to say, okay, is this really marketing hype or is this marketing reality? And if you are not using the tools like this, are you bringing a knife to a gunfight? Uh, and, uh, there's, there's a real question as to is it going to be ethically, you know, we're gonna talk about ethics in a bit, but there's a real question. Is it ethically? Uh, are you not doing your ethical duty if you don't use these tools that maybe by declining to use the tools, maybe that's the ethical problem,

Right? Perhaps if we take a step back and consider, okay, if we don't use a pen and paper, are we really doing the best for our clients?

That's right. There's, there's been all sorts of judicial opinions, uh, saying just that, uh, like someone would had asked for fees, uh, if they, when they had done research on, on books, and the, the, the judge said, no, you don't get your attorney's fees because you wasted time, because you didn't fulfill your ethical duty to do legal research like God intended to do electronically.

There we go. Now, Damien, how is AI reshaping the skills that junior lawyers and law students need to develop to remain competitive?

I think that, you know, I, when I was a baby lawyer in, uh, 2002, uh, I, I thought, you know, if I just, uh, research really well, and if I write really great briefs, of course I'm gonna make partner. Uh, that's what I, in 2002, and of course that was not true, uh, because the people who made partner were not just those who wrote briefs and did really well, uh, with their legal work. Of course, that was all well and good, but the people that made partner were the people that brought in money. That is, if you brought in the clients, and if you were able to not just do good work, but bring in enough clients to bring in cash, um, I know really horrible lawyers, uh, that really drove, drafted horrible briefs that made partner because they brought in money. Uh, and so, so that is the dirty secret of law practices.

It doesn't matter how well you do it in your practice, uh, in your legal work, uh, really the, uh, in the end law law firms are businesses. And so when we take it through, look at it through that lens, um, AI is now a way that we can maybe make more money as businesses than we ever have in the past. And the way that we can do that is, um, in the old world, if you're billing by the hour, um, you would say, well, wait, why would I pay for a tool that makes me less money? Uh, because I, I get paid by the hour and, and the slowest lawyer wins the race. Because if you spend 10 hours on a thing that I spend one hour on, you make more money than I do. Um, and that's, uh, under the bill hour paradigm, uh, that I'm not incentivized to do things faster.

Uh, but if you maybe move away from that billable hour paradigm to go with my, with my friends, a lot of flat fees, and with flat fees, you, if you shrink your cost, you increase your profit margin. Uh, and so this, uh, this flat fee paradigm or, uh, this, uh, this subscription model to say, Hey, as a small business, you pay me x amount, number of dollars a month, uh, and then I'll give you, uh, x number of hours. And my friends who do that, they say, this is just like private equity that really asks about the startups. They say, how much is your annual recurring revenue? This is solid monthly bills that you get every single day as revenue. It doesn't matter if you have a good month or a bad month, you're getting the same subscription fees. So if you move away from the bill hour and you move over to flat fees and subscription fees, you can actually make way more money because you can use AI to shrink your costs and take that 12 hour task and shrink it to one minute, and then you can do sell more business.

And that is what the junior associates, when I was growing up, didn't realize, is that making money is the thing, <laugh>, we are a business after all. Uh, and so I, I think that we, um, we as junior associates, um, can do that for a law firm, uh, or you could jump ship and be able to start a solo small practice. And there's never been a better time to be able to be a solo small lawyer. Because if you are either associate or a, say a partner at a big law firm, and you see the light to say, oh, I'm going to, um, go away from the Bill Lauer practice, and I'm gonna instead do the flat fee, uh, their partners at the big law firm might say, no, I'm making a lot of money with the Bill Auer, thank you very much. We're gonna stay billable hour.

But if that person, that partner jumped ship from the big law firm, and then, uh, has a solo firm, they can say to their clients, I'm the same person I was 20 minutes ago. They bill by the hour. I'm gonna do flat fee and I'm gonna pay, I'm gonna charge half of what they do, and I'm still gonna make more money, and then they can get more clients. So really, there's a, there's a big question as to how much the big law firms, if they don't go to the new models, that is the flat fee and the this, uh, the subscription fees and things like that, success fees, contingency fees, um, I, I wonder how much the big firms will stick around. And we can talk about the type of work. I know that we're running short on time, but, um, there's cream work at the top that is the most, uh, the most, uh, bet the company work or the, the biggest m and a. Um, and then at the bottom is commodity work that really is, is, uh, low. But in the middle of that is core work. And the real question is, how much do your clients come to you for the cream work? And really, how much is core and commodity, and how much of that will be taken by af?

Um, so what is the current ratio that AI is really, you know, um, efficient at taking over? Is it the cream, the commodity? Is it the core?

So, uh, a friend of mine, Jay, um, is, uh, is one of the smartest people. She worked for Baker Mackenzie, um, one of the biggest law firms in the world. And she did pricing for them. And she estimated that the amount of corporate legal spend, uh, for cream, uh, was only about, uh, 15% of corporate legal spend. Um, and, uh, I I, I'm, I'm gonna get, I'm gonna get this wrong. So I'm, I'm gonna actually look this, uh, look this up. Uh, the, um, it's, uh, it's corporate, uh, no, uh, 15% cream that was right, uh, 5% commodity. 'cause they've already pushed down the cost of that to push it to low cost, uh, alternative wheel service providers and that kinda thing. So 15% at the cream, uh, uh, 5% commodity. And then 80% of the corporate spend is in core. And this is not commodity things, but it's stuff that's, you know, just run of the mill litigation or, you know, just regular contracts that is the core.

But then she estimates how much can AI take, and AI will of course, take 100% of the commodity, um, and probably 85% of the core and maybe 50% of the cream. And so if that's, that's an estimate, but I think it's a pretty good estimate, and it, but it's, that's true. Then, um, how can you run a business on half of the cream and only 15% of the core? Um, I think that, uh, law firms need to be able to think a lot about, uh, where they're gonna be getting their money. And one other thing is that, um, in the United States, at least one 92% of legal needs are unmet because we lawyers are too expensive. And so I wonder, uh, we as lawyers should be thinking, um, that 92%, um, economists would call that a latent market that is just waiting to be tapped.

So what if we take our AI shrinking our costs to increase our profit margins? What if we take those lower costs and reach into that 92% and be able to tap into that market? And maybe we can make more money than we ever have, and we can actually serve access to justice in a way that justice has not been served before. So I think that we, um, we are really need to think as a profession about where we have been making money and where we will make money. And I think those may be two very different places.

That's amazing. That's amazing. I think, um, you know, that's truly enlightening with regards to the real skills that junior lawyers and law students need to focus on to remain competitive. Uh, now, Damien, are there any traditional training experiences or early career learning opportunities that risk being lost due to increased automation? And how can the profession address this?

So let's think about how training happened in the old world. Uh, so when I was growing, uh, up as a baby lawyer, I would spend hours and hours with bankers boxes full of paper, looking through documents and trying to find that smoking gun from a bunch of boring documents and the number of smoking guns that I would find in, in a day's time in maybe one smoking gun. Uh, and the rest of those eight and a half hours were just wasted looking through boring nothingness. Um, the dirty secret though is though, uh, my, my 12 hours of reading for one smoking gun, that 12 hours of reading made my law firm a lot of money. Uh, and so really, um, uh, that question is, was reading for 12 hours boring things for one smoking gun, making me a better lawyer? Was that training me? And the answer was no.

It was just wasting my time, uh, and wasting, uh, the client's money. Uh, it was not making me a better lawyer, but really that's in the old world, what we really, that's what really made itself out to be training, is to be able to just throw bodies at a, at a problem and let them churn and maybe build a client for as much as they want. So right now, I think that we now have an opportunity to train our lawyers better because people went to law school to be able to consult with the clients and to be able to help people with their problems. And if we are be able to shrink our costs to reach into that 92%, um, then we can maybe reach into, have the associates actually do more one-on-one communications with the clients, maybe the lower income or middle income people that couldn't afford it before, and then they could actually be able to get more boots on the ground training to be able to solve business problems or to solve litigation problems, or to solve advisory problems or regulatory problems.

Um, that's, that's way better training than me reading bankers boxes in the old world, uh, where I didn't, well, that wasn't really training at all. So really, I think that with ai, we could be able to get rid of the drudgery. Um, my friend Casey Flaherty says we do, we as lawyers do a lot of thinking, and we do a lot of thunking, uh, we do a lot of drudgery, uh, and maybe AI can do a lot of that thunking for us and just leaving us to do thinking and leaving us to talk to our clients more and leaving associates to be trained more by actually getting more of that work. So I, I think that people say that, you know, our our associates are, um, are not gonna get the training that they have been. I, I think are just mistaken. I think that we, um, if they are, if we reach the right client pool, that is the, the middle class that isn't being served, and we, if we price our, our, uh, services well enough to reach that middle class, I, I think that we will get plenty of training for the next generation.

Mm-hmm. And hands on at that too.

A hundred percent.

Okay. So that was very interesting because now, uh, when you break it down like that, uh, do you believe that primarily in the traditional, um, the old world training, it was more geared towards training, uh, cases based for, you know, the people who can afford it, those, that 8%, uh, and leaving the rest of the 92% in the dark?

A hundred percent. So essentially, rich people were subsidizing rich people. And there's a, um, uh, a joke, uh, an open joke in my law firm when I was an associate is that we as associates could not afford to hire ourselves. That is, we couldn't afford our own billable rates. Uh, and if we couldn't afford as upper middle class people to hire ourselves, like what choice does the middle class person have? And so I, I think that, uh, yes, we, uh, we have historically been rich clients paying for rich lawyers to be able to do their work. And, uh, at least in the United States, there's a lot of discontent in the populace saying, the system is rigged against me. And with that number of 92%, maybe they're right, maybe the system is rigged and not on purpose. You know, I think our judges and our judicial system do the best that they can, but the system is just not really set up to serve that 92%, or at least it hasn't been until now.

Now, AI can be able to shrink legal fees so that I, as a poor person can use chatt PT, and it gives me pretty good advice. It doesn't give me yesterday's case. It doesn't give me yesterday's statute, it doesn't give me yesterday's regulation. Uh, for that you need something like Vincent. But if you're looking for good enough, AI is pretty good for good enough. And so now we have to say, okay, what do we as lawyers do with chat GPT and Gemini and Claude being out there doing this work? Maybe we can compete with them by giving better service. So I, I think that this is maybe a golden age for lawyers. We can serve that 92%. There

We go. Uh, now, Damien, what are some of the new expectations that clients have regarding speed, cost, and transparency as AI becomes more integrated into legal services?

Uh, I'll answer that with a, a true story that, um, I was on stage with Salesforce, uh, one of the biggest companies in the world, and, uh, Salesforce at Ilta Con last July said that in the first six months of 2025, in those first six months, they saved $3 million in legal fees that they used to spend on law firms, that they brought inhouse using tools like Vincent, that was $3 million in six months, and they expected five or $6 million by the end of the year that they used to spend on lawyers, that they're using tools like mine to not spend money on lawyers. And so, um, you know, a lot of, uh, a lot of lawyers just don't know why the phone stopped ringing. And so I, I, I hear that from customers, uh, clients over and over again to say, you know, for basic questions about what does the law say about this?

Or does this violate EU law? Or does this violate French law or Spanish law, um, or does this file in la Latin American law or Singapore law, um, those kind of questions just don't get asked anymore. And the lawyers can point their fingers all over the place to say, oh, it's the bad economy, or it's, the client had a sweetheart deal. But the point is that they're just not need for what they're giving. And so my, my friend Darth Vaughn, who works for Ford Motor Company, uh, as an in-house lawyer, uh, he's litigated for a bunch of years, and then he went in-house at Ford Motor Company, he said to a bunch of lawyers, uh, law firm lawyers, uh, he said, Hey, law firm lawyers, you have a new competitor in town. You know who, who it is, it's me as, as Ford Motor Company, and you have to tell me what you can do that I can't do as Ford Motor Company within my walls, because I already have a bunch of AI tools.

I can already vibe code and do a lot of things internally. So you have to tell me what is the value that you're giving me above what I can do myself that's worth me pulling out my hard earned dollars and wallet and giving to you. And so I, I think that, that, um, it's not just doing work faster, better, stronger, but just showing what value are you providing as a lawyer above what in-house counsel are doing? And I think one answer, uh, to that question is what we talked about just a bit ago is trust. That is a large corporation hires you not just to be able to do the widgets, uh, of the, you know, drafting the motions or drafting the documents, but also for your, what's stuck in your brain. Uh, because I, as an in-house counsel say, I want to know, uh, that this judge hates it when you do that, or that if you, we can't argue this at the negotiating table for the m because that'll screw, that'll scuttle the m and a deal.

That's stuff that's stuck in lawyers' brains, that it's worth me as a corporate lawyer to hire you, uh, because I don't have that insight of what the judge says or what will not go well in the, the negotiating table, or what happens when you represent my competitors. And what happens in deals like these, all of those aspects are stuck in the lawyer's brain in the way that's, I just as an in-house counsel don't have access to. So, and lawyers really have to think about, um, that what's stuck in your brain is your moat. That is, uh, small businesses are always thinking about, what's my moat? What keeps my competitors at bay? And your moat as a lawyer is what's stuck in your brain that is not yet in an artifact like a document that AI can ingest. So as long as it's stuck in your brain and not stuck in an artifact that AI can adjust, that is your moat against the ai.

And that is your moat, uh, for the in-house counsel, uh, because they need that stuff stuck in your brain. But the real question then you have as a law firm lawyer, is how many of your competitors have taken that stuff in their brain and offloaded it onto an artifact and then made that into a product that people buy? And how many things that are stuck in your brain are truly unique and truly have enough value that it's worth Ford Motor Company and others pulling out their wallet and paying for it? And I think that's the real question as we go forward, is, is what is our remote, uh, what is our competitive advantage? And, uh, how soon can we be able to, uh, keep using that, uh, before it gets commoditized by others?

Uh, now Damien, what are the most significant ethical and risk concerns lawyers should be mindful when using AI in legal practices today?

Sure. Everyone thinks about the hallucinated cases. That is, you use chatt PT to be able to do the research. Uh, it cites hallucinated case that makes its way into the brief. Uh, and then the judge sanctions you for, uh, doing the hallucinated cases. Um, that, um, that is what everyone, you know, I think there's been over 150 or maybe 200 of those cases. Uh, so everyone thinks about those ethical risks. Um, there's a good solution to that. Um, and my friend, uh, and colleague Ed Walters says that solution is read the damn cases easy, right? Uh, if you just read the cases, you could be able to avoid hallucinated cases. And that's something that we've done. You know, speaking of the olden days, right? Uh, we would always read the cases, right? Uh, so, uh, so I think that, um, you know, the ethical risk of hallucinated cases is really a non-risk.

If you are a competent lawyer that has done what you were supposed to do is to read the case, uh, for which you're citing for that particular proposition. So everyone thinks about hallucinated cases. Um, another, um, aspect of ethics is to think about, um, not just, you know, hallucinated cases, but, um, we mentioned this earlier, but what is your ethical duty to charge a reasonable fee? And how, how that is something that, uh, it is unethical to charge a fee that is too high. And the real question is, if I spend 12 hours doing that legal research that a tool like mine literally takes five minutes to give you a better work product, is that a reasonable fee or not? And at least for legal malpractice in the United States, um, the question of whether something is reasonable or unreasonable is gauged by people in your peer group that is, people in this city and this type of practice.

Would lawyers like you, uh, do it research in the olden way or the new way? And so a bit of this is kind of everyone waiting and seeing, well, is everybody else gonna run or am I gonna run? Uh, but the faster everybody else runs, the more you at risk, uh, of being and, uh, behind your ethical duties because everyone else is doing things the new way and you're doing things the old way. So everyone talks about the ethical risk of using ai. I think we should be saying, what is the ethical risk of not using ai, uh, and being able to charging your clients too high of a fee? Uh, that's, that's number one. Uh, number two is, uh, there's a real risk of losing business. And that, uh, you know, don't, people don't think of that as an ethical risk. Um, but I think the ethical risk of getting, uh, dinged for hallucination, sanction is very, very slim.

Uh, but the business risk of losing money because your clients just don't need you anymore, <laugh>, that's, that's a much higher risk. Uh, and so, uh, anyway, so, uh, and if you don't use ai, um, then you've avoided that ethical risk, uh, and also lost your business. Uh, so there's a real, a real question as to, um, you know, people that don't use AI because of the ethical risks, I think are misplacing the opportunity. Um, so, uh, I I think that, uh, regulatory risk, um, I, I, I don't know, I don't know. I'll take this a little bit differently. So the, the regulatory barrier for a non-lawyer to be able to practice law in the United States is the unauthorized practice of law, uh, that if you're a non-lawyer, you can't practice law. But I know a lot of companies that are not giving legal advice, but they're instead giving legal information.

Chatt PT gives legal information. What are the cases? What are the statutes? What are the regulations say? That is not legal advice. I'm not telling you what to do, I'm just telling you what they say. And that is not practicing law, that is just giving legal information. And so I think that the regulatory moat, uh, that says that only lawyers could be able to provide legal services. There's a real question of how much of the legal services that lawyers provide is truly legal advice. You should do something versus legal information. This is what the case, the statute and the regulation says, because if you're just giving legal information and not legal advice, that's the same as what Chad JPT can do to say, this is what the case says, but the statute says, and are you really practicing law if you're merely providing legal information?

And, uh, if you'll give me 10 more seconds, uh, I, I thought about, um, this, when I was a, when I was a litigator, I thought, uh, I had the revelation about a year ago to say, how much of my 15 year litigation career was I really giving legal advice versus just legal information? Because when I would give legal information, I would say, uh, you have three options. This is what the case statute and regulation says. Here are the three options you have for you. And the client would ask me, which one should I do? And I would say, oh, that's a, that's a business decision. I would say, uh, what's your appetite for risk? Uh, if you're really risky, you do two and three and one, uh, but that's your decision, not mine. And I realized about a year and a half ago, I didn't give any advice.

I was merely giving legal information and I was giving options. And the reason I didn't give advice is because of, uh, you know, subconsciously that was, that was a malpractice risk, because if I say, you should definitely take number three and you take number three, and bad things happen, and I could get sued for malpractice, but if I say it's a business decision, I've offloaded that advice part, and I just gave information that saved me on the malpractice side. But was I really practicing law? And so I, I think we really need to think about if we're gonna be competing with the ai, maybe we should give more advice and give the more humanity in the way that the chat GPT and others can't.

Now, um, what about client information, right? Uh, how, how does that play into the, uh, ethical standards of things?

So, uh, there is, uh, a, a joke, but it's not a joke that, uh, if you're not paying for a product, uh, you are the product, uh, you're not the customer. Uh, and so if you are not paying for chatt PT or you're not paying for Gemini, um, there's no guarantee that they will not train on your data. So if you upload your client data, they may well train on it. Um, does that violate ethics? Uh, I'll let you talk to your regulator, but a lot of jurisdictions say, yeah, that would violate your ethical obligations. But if you were to instead, you know, uh, the tool I'm building with my team is Vincent, uh, that, uh, that has agreements saying that we will never train on your data. Uh, that's point number one. And number two, our partners at OpenAI and Anthropic and Google will not train on your data.

Uh, there's zero data retention there, and we pay a lot of money to make that true. And then number three, to say, if your data is uploaded, it is secure because we have SOC two type two certification. We also have ISO security certification that is third party auditors looking through our systems and saying, your data is secure at rest. So if your data is secure at rest, and we won't train on your data and our competi, and then our partners won't train on your data, um, that is just like, uh, you know, do I trust Microsoft to not train on my data when I've been using Outlook for the last 30 years? And the answer is, of course, because we have a contract with Microsoft saying that they will not, uh, breach our confidentiality. And if that's true, that contractually, you're now able to upload your client's safe data to email, maybe then contractually, you can also upload your client's data to Vincent and to other tools like it.

So I think that we are really, um, uh, when I was back in 1999, um, there was a real question as to whether it was ethical to be able to email our clients, because email is inherently insecure because it's, uh, if it's unencrypted, you couldn't eavesdrop and listen to email. But of course, everybody used email and, uh, and it was a real question, can I go to the cloud? Uh, you know, is that ethical? Everybody's in the cloud, right? So we know where the movie ends. Uh, is it ethical to upload your client's data? Of course, it's 'cause you have contracts.

Now, Damien, who should ultimately be accountable for AI assisted legal work, the individual lawyer, the firm, or the technology provider,

<laugh>, that's, uh, there's a, uh, in the United States, uh, that we had a president named Dwight Eisenhower, and he said, the buck stops here. Uh, that, that, uh, that he said that I'm the US president. Uh, therefore any decisions that are made under me, I'm gonna take responsibility for it. Uh, so I think that, uh, the answer to your question is, uh, whoever signs the piece of paper, whether that's a contract or whether that's a brief pleading or motion in the United States, we have something called Rule 11, uh, saying that, uh, when I sign a document, uh, that means that everything above my signature is true and correct, and it's true and correct whether a paralegal wrote it true and correct, whether an associate wrote it or it's true and correct, whether it was drafted by an AI under any one of those, uh, it's my signature, therefore it's my being bound by Rule 11.

And so I think the bus buck always stops with you because it's your professional responsibility to tell the court or to tell opposing counsel, uh, and be straight and true with them. Um, and so, uh, of course, uh, I as a lawyer might, uh, have my law firm's malpractice insurance covering me, right? So to your question is, is it stop with a lawyer or does it stop with the firm? Well, is the, if I, if I commit malpractice, uh, is the firm gonna pay for my defense for malpractice? Uh, at least the firm's insure, then I guess maybe in a sense it would be the law firm where the buck would stop. But then, uh, the third question is technology providers. Um, there's a real question, like, if I used Westlaw wrong, could I sue Thom Reuters for giving me bad results and not giving me the, the case that I wanted in the first page when it was actually, uh, stuck in the fourth page?

Uh, is that, uh, can I soothe Westlaw for giving me bad results? And the answer is no, of course not. That is a tool. Uh, your job as a lawyer to is to use the best tool for the job. And just because you didn't use the tool well enough, uh, doesn't mean that the tool was bad. Uh, and so really there's, uh, there's a question as to, you know, if, uh, if, uh, chat BT hallucinates a case, can you sue chat BT? That seems silly, right? Because you know that there's a risk of hallucination <laugh>, so why would you sue OpenAI, which makes chat pt? So anyway, so, um, so between those three, between the lawyer and the law firm and the technology provider, uh, the buck always stops with the lawyer, uh, and maybe with the firm if they cover the malpractice.

Now, Damien, what about malpractice insurance? Uh, are there any updates that, um, lawyers should be looking for right now?

Uh, malpractice insurance, uh, I, I don't know of many malpractice insurers that have spoken about AI recently. I know when Chatt PT first came out in 22, 0, 23 or so in March of 23, I know one large malpractice insurer in the United States, um, said, Hey, beware of artificial intelligence. And I think they were largely looking to, um, right around that was right around the time that Steve Schwartz was the first AI lawyer that gave hallucinated cases. Uh, so one, uh, insurance malpractice insurance carrier in the US said, Hey, don't be like Steve Schwartz <laugh>. Don't, don't do, uh, don't do, uh, don't do what he did. Um, but that was three years ago, uh, almost to the day, I think it was March of 2023. Um, but I've not heard of any malpractice insurers talk about it since. Um, largely because a lot of the problems that were there in 23 are solved problems now, at least with tools like mine, where we, uh, we give you the actual non hallucinated text from the actual non hallucinated cases, 100% of the time, it's not 90%, not 99%, but 100% of the time the text on the right hand side will be non hallucinated.

And 100% of the time the case will be non hallucinated. And you might say, well, Damien, how can you say a hundred percent? And not 99%? And I can say a hundred percent is because this text was not created by a large language model, was not created by ai. It was created by a database search, just like we've been doing database searches for 40 years. So 100% of the time, if that text and case are non hallucinated, it's a solved problem. And really, uh, so that's why I think malpractice insurers haven't been talking about it recently.

Let's look at what this future ready lawyer and law firm look like in an AI driven legal ecosystem.

I think, you know, yeah, a Future Ready lawyer is, uh, there's a, uh, a hockey player named Wayne Gretzky that said, uh, I don't skate to, I don't skate to where the puck is. I sc skate to where the puck will be. And so I, I think that if you really wanna be future ready, you have to say what's coming down the pike. Uh, everyone's talked about hallucinated cases. Everyone's talked about, you know, I've used CHATT PT to be able to do my marketing material. Uh, that's not future ready, that's just today ready, or actually yesterday ready. Uh, if you really wanna be future ready, um, you have to think about what's coming down the pike, where's the puck going? And where the puck is going is that, uh, we, um, are increasingly in an age agentic world. And, uh, if you don't know about agents, it's that, um, there's a, it's a good example is that, uh, open claw, uh, is a new paradigm where you have hundreds of agents within a machine all doing work, and they call them agent swarms.

So you have these artificial intelligence bots going out to your calendar and seeing what you need to be doing today, and they go to your email and say, here's some drafts responses to the emails. And then they go, if you wanna go on a trip, they'll go to all the websites and they'll see, okay, what are the things to do? What are the best deals on those things? And maybe they make a phone call to be able to see, you know, what, whether they can get a better deal. So these are all agents that are working on your behalf without you doing a thing. And so, um, thinking, uh, about where Future Ready lawyers are going to be, um, they're probably gonna be thinking about, the thing that I've been thinking about and been talking with my team about is that right now we're building, um, a swarm of agents, uh, but they're lawyer agents.

And what lawyer agents, uh, one of them is an associate agent who does what I did as a, as an associate at a law firm, where I would say, here's all of my clients' emails, all my clients' documents, all my clients' Slack messages. And the, uh, associate bot will go through those and we'll do issue spotting to be able to say, uh, well, this looks like a breach of contract issue, or This looks like a trade secret issue. Uh, right. So it does the issue spotting to identify causes of action. And then it also identifies what are the jurisdictions that are applied? Is this New York law or is this California law? And then it does research to be able to say, what are the elements of breach of contract in New York? And what are the elements of breach of contract in California? And then once it has those elements, then you could say, okay, based on the documents I have, can you fill in buckets one, two, and three in California?

And can you fill in buckets one, two, and three in New York? And what if bucket two is missing? There's no evidence for bucket two. It will draft questions for the client to be able to say, Hey, client, can you fill in bucket two to answer this question? So it'll do that for a while and do all that work that an associate lawyer would do. But then we also have a partner bot that then criticizes the associate and says, well, have you thought about this? Have you thought about that? And then the associate says, you're right, and then goes off and do this thing. So we'll do some back and forth then, but then we also have an opposing counsel bot that we then throw the motion or the brief or the pleading over to the opposing counsel, and then they respond saying, no, you're gonna lose because I'm gonna argue this, that, and the other thing.

And then the associate bot on the other side gives it up to the partner bot who will then throw rocks, right? So they go back and forth and then take the final product from them and give it back to my bots. And then I can make my, my arguments better, because I can cut those arguments off at the past and bolster their arguments. But then we also have judge bots where then we throw the outputs into the judge, and the unbiased judge is able to say, uh, I'm gonna rule for you because of this reason, but it's not just one judge. It could be three judges or five judges. And then we could say, 60% of the time, the judge loves this way, but 40% of the time the judge loves this way. And then take that output. And then my associate bots can do even better to be able to draft, to be able to anticipate to say, judges like this more.

So I'm gonna put this in there. Everything I just showed you or it spoke to you about was before any human even touches it. So after all of that work, you get the end product to be able to say, now this is a batten by opposing counsel. Uh, it was, it was, uh, you know, thrown rocks at by the judge. This is maybe the statistically likely argument. And then you as a, a human can say, perfect, let's get it out the door. Or you could say, Hey, associate bot, maybe you should do this other thing and then start the whole process again. So this is where the future Ready lawyers thinking about this isn't just chatt PT what I described. This is agent swarms, lawyer swarms that are doing better than any human can do. And the real question is, what are we teaching our associates, uh, to do in this world? And what do law students should, what should they be thinking about in this world? I don't know if anybody knows.

Wow. Uh, yeah, that's just complete simulations you were just talking about. That just happens in the background. And, um, I mean, I'm just wondering, what, what would a lawyer on the opposing counsel who does not pick up these agentic swarms, really, where would they stand if they're, you know, facing you <laugh>

That's, you know, it's like bringing a knife to a gunfight, you know, if, if you're not doing this, um, I, I was, um, literally my next door neighbor, uh, over here wants to build a, a fence, uh, right up against my house within six inches of my house. Uh, and so, um, I, I've used Vincent to be able to find really good arguments, uh, to be able to say, no, I have an easement, or I have adverse possession, or I have practical boundary. Uh, there are all sorts of causes of action. And I very quickly created many pages of really, really good illegal arguments that are all grounded in actual cases, actual statutes and actual regulations. The opposing counsel, uh, for my, my neighbor, uh, accused me of having AI slop. And, uh, then AI slap, if you don't know, is just like stuff that AI is just cranking out, like marketing materials.

But I was offended by that because I, I said, when I worked at a massive law firm, a Robins Kaplan, if that massive law firm cranked out 25 pages of solid legal argument, that's really good lawyering. But because I used AI to crank out 25 pages of even better lawyering than I did at Robins Kaplan, but I used AI to do it and to help me, that's AI slop. Hmm. Even though it was better output than I did at Robins Kaplan, those 25 pages, is it AI slop or is it just the way lawyering is done today? Yeah, and I think that that, um, bringing a knife to a gunfight is not a point of honor that is just stupidity. And if I, if you say, oh no, I do all of my research in books, thank you very much. Do you respect that lawyer more? Of course not because they're doing it wrong. Or if they say, the lawyers that say, I don't use email <laugh>. I don't have a cell phone. Do you respect that lawyer? So really, when people say, I don't use ai, how is that gonna weather in 10 years, in five years next year? It's like bringing a knife to a gun fight.

Absolutely. Outdated technology right now, Damian, looking ahead to five to 10 years, how do you expect AI to really reshape the lawyer client relationship and the delivery of those legal services?

Um, there's a, there's a, uh, saying that the, I dunno if you've heard of Ray Kurzweil, but he, in 1992 said that the singularity, that is the point where, uh, human intelligence and machine intelligence would, would meet and the machines would blow by us. He predicted back in 1992 that that would be in 2030. Uh, and, uh, and he actually, uh, was a little bit late on that because, uh, AI has already blown past, um, most people in intelligence. So whenever somebody asks me a question like you just asked to say, where are we gonna be in five years? Uh, man, I don't know where we're gonna be in six months. Uh, because, you know, if you'd asked me a year ago whether we would be here today, that is AI beating, uh, 99, it got 99% of the questions right? Or actually a hundred percent of the questions right on the bar exam.

So in 2003, it beat 90% of humans, and today it beats 100% of humans on the bar exam. Um, so if you'd asked six months ago, tell me about agent swarms, I would've been like, you're crazy. We're never gonna be there. Right? But here we are, we're the thing I just mentioned, is now definitely possible. So, um, a year from now, gosh, I have no idea where we're gonna be five years from now. Oh, man, uh, like we're, we're gonna be going maybe over some over 10 window that I just don't know what's on the other side. Um, but I'll give you a little bit of, um, in November, uh, I was asked to go to the Vatican to talk about AI with the Vatican. And, uh, while I was there, I was working with, uh, uh, and col collaborating with, uh, somebody who, uh, works on Kate Ray Kurzweil's team.

And we were talking about what happens when maybe the cost of legal services shrinks because the, uh, the number, uh, 12 hours gets shrunk to two minutes. And it's not just legal services, but also accounting services and also, you know, any consultancy. Uh, and so maybe the cost of services just shrinks. And then as robots do more work and robots are doing more work, maybe the cost of goods similarly shrinks. And when you have the cost of goods and services, the economy wide shrinks, then you end up with massive deflation, not inflation, but deflation. And what does a world look like with massive deflation? Uh, and, and yours? Uh, you know, one path is that we, as lawyers make way more money because we can shrink our costs and be able to reach what's called, uh, Jevons Paradox. Uh, and for people who don't know what Jevons Paradox is, uh, we thought that, uh, Jevons Paradox means we thought LED light bulbs were gonna save us a lot of energy, but they're so cheap, we just leave the lights on all the time.

Jevons Paradox says that if you, uh, make something more efficient, society uses more of it. And so, uh, right now, light bulbs are too cheap to measure. And what if legal services were too cheap to measure? What if, uh, with Jevons Paradox, what if you would say to me as a lawyer, you'd say, uh, every time I call Damien, uh, it's a thousand bucks, forget it. I'm gonna, I'm gonna risk it, right? But if that thousand bucks shrinks to $50 or $25, I'll call you every day. In fact, maybe I'll spend more than a thousand bucks in a given year. Uh, so maybe if you shrink the unit cost, people spend more money on things just like they leave the lights on all the time. So that's a long way of answering your question. That five years from now, who knows, right? Maybe we just have a shrinking deflation in the economy.

And, uh, when that happens, who knows? Um, or maybe we have Jevons Paradox where we can actually use that shrunk cost to serve that 92% of legal needs that are unmet and be able to have that 92%, maybe we can dip into that latent market and serve all the people that haven't been served. So anyway, so if, if you were to gimme a crystal ball, I think that we're gonna serve in five years more people, uh, than we have been in the past. And I think everyone will have a better idea of what's their legal obligations are and what they should do legally. And I would have to say that's, uh, that's probably a better world for our society. Uh, last thing I'll say is that my friend Bridget McCormick would say that, uh, she's from the American Arbitration Association. Uh, she says that law is society's operating system. Windows is the operating system for machines. Um, the law is the operating system for society. So I think that, uh, five years from now, our operating system is gonna be doing pretty well.

Wow. Okay. Yeah, I was just thinking perhaps it's about time we meet that investment of technology and finally see that market correction. Um, there's always that question about is AI and data centers hiking up electricity prices? And we're all witnessing that all over. Um, but perhaps that's for another time.

I, I, I really outside have me back. I'm happy to talk about such things.

Wonderful. So Damien, um, if, if you had to summarize this entire hour into one sentence, for our lawyers listening in, what is the headline?

The headline is, jump in and start using the tools. Uh, because I, the people say, well, what books should I read? Who should I listen to? And that's almost as silly as saying, what book should I read to learn how to swim? And the answer is, of course, you don't read a book to learn how to swim. You learn how to swim by swimming. So I would say that the one takeaway is to, if you've been intrigued a little bit by what we've been talking about for this last hour, um, jump in and use the free tools. That is the chatt PT outta the box. Uh, so that's number one. Number two, uh, spend the 20 bucks a month to be able to see how much better that tool is and it's, it's way better. And then learn how good or bad those two are. And then after that, spend a little money on a tool that is built for lawyers, by lawyers that has yesterday's case and yesterday's statute and yesterday's regulation, and see how much better those are than the first two. So I would say, uh, start swimming and see, uh, how, uh, you should really bring a gun to a gunfight

And what a remarkable conversation that was. A huge thank you to Damien Rio for the depth, the breadth and the honesty he brought to this episode. From the death of the Billable Hour and the rise of subscription and flat fee models to the truth that 92% of legal needs in the US are unmet because lawyers are too expensive to the agentic future of legal work where associate bots argue with partner bots and judge bots stress test the brief before a human ever touches it. There is so much in this conversation that every firm needs to sit with. And that brings me back to our sponsor, Equa Marketing, because everything Damien talked about today, from AgTech workflows to reaching the latent market of underserved clients. All of it depends on one thing. People have to be able to find your firm. Equa builds the digital visibility and intake systems that make sure when the right client is searching, your firm is the one that shows up. Their team will sit down with you, audit your online presence, study your competition, and walk you through custom recommendations. That session is normally $900 and it is completely free for our listeners. Book it at www.dominantlaw.com/msm. If this conversation got you thinking, please share it with one attorney who needs to hear it and, uh, follow the Dominate Law Podcast. So the next episode lands right in your feed. I'm Donna Deisha. Thank you for rebuilding the law firm of the future with us. We'll see you in the next episode.