Learn from the experts behind the growth of sterlinglawyers.com Anthony Karls, President of Rocket Clicks/co-founder of Sterling Lawyers, and Tyler Dolph, CEO of Rocket Clicks, interview the experts in all the areas that will drive revenue and increase profits for family law firms
Get technical knowledge and learn from the experience of those who paid the price to learn what it takes to grow from an idea to an exclusively family law firm with 30+ attorneys.
Law Firm AI Policy: Safety Is A Tier Issue, Not A Tool Issue
Consumer Vs Enterprise AI: Why Free Accounts Make You The Product
The Silent Killer: Why Liability Rolls To The Firm Owner
ABA Opinion 512 And State-By-State Ethics Guidance
AI Trial Prep And Drafting: 100 Questions To 300 Angles
Inside Sterling's AI Governance Policy Build
Is the ethics fear keeping you from leveraging AI? It's justified, just not for the reason you think. Here's everything you need to know about ethics and policy in AI. For family lawyers, welcome back to the Sterling Family Law Show. The show for family law firm owners who want to grow their profits, multiply their cases and set a clear roadmap on what they need to do to build the firm of their dreams. our episodes breakdown proven business growth, digital marketing, and revenue for strategies to help you scale smarter and not just be busier. I am your host, Tyler. I'm also the CEO of our hyper focused digital marketing consultancy called Rocket Clicks that was born out of our very own family law firm, Sterling Lawyers, and today we are talking AI. I have Tony Carl's, who is the co-founder of Sterling Lawyers with us, and we are going deep. Today's episode, we outline the false belief that AI is not safe for family lawyers and why safety is a tier issue, not a tool issue. We're going to talk about the formal ABA opinion on AI, and then we're going to talk about building your own AI policy, something that every firm who wants to use AI needs in place. Tony. Welcome back. We're getting deep into the weeds with AI and family law firms. Today, I want to start on the false belief that I feel like some family attorneys have, which is that AI is not safe for law firms, and I think maybe done incorrectly. That's correct. It's totally not safe. But there is a way to leverage AI for your family law firm. Talk us through the difference between enterprise, you know, accounts and AI versus like your run of the mill chatbot that people download on their phones. Yeah. Most of these platforms plod OpenAI Z with grok, and then Gemini. They all have essentially there's two streams of interaction that you can have with them. One would be what's called a consumer grade account. And those are all you can get them in a couple different variants, but typically you can get all of them for free. So you can go sign up right now and just put an email in. And then you get access to use the LM chat with it. You can do certain things with it, but it's not free. Free means there you're the product. So just know that. And if you're a lawyer and you're the product, the product, that means your data is the product and you can't be sharing that information with with these companies because they're going to use these. These are going to be the data is going to be used to train their models. And if you're putting client data in that's now part of the public repository. And you you have actually violated ethics rules. So so there's that vein. And it doesn't matter if you're in a if you're in a consumer grade product and it's free or you're doing like a plus version where you get a little extra utilization or you're doing a max version or you're getting the whole kit and caboodle, it's still a consumer grade terms of service. And like you got to read through it. They're going to say they're going to say that you have some control over whether it turns on or off and certain things like that. But the reality is, is like it is a different terms of service. It's a different contract, it's a different workflow. And it is piped for leveraging your data to do data model training. And it's even the paid versions I would say are risky to use, and you should be entering a contractual relationship with them in what would be called an enterprise terms of service. And again, on that side, there's multiple different versions of that. What I'm really talking about is your how you're interacting with them from a terms perspective. When you enter a business contract, those that data is then contractually obligated not to be used for creating more or used for training. You have you have more controls. There's more features from a business perspective so that you can do it. Typically they come in a couple of basic janas. You have the teams, which is kind of your entry level account. Same same contract but lesser tools. You have an enterprise version which is going to have a whole bunch more tooling, but it's the same contract as the team account. And then you have what's called a platform account, which is really your API. If you're building applications directly to these LMS and you need to plug, you need to plug the application into the AI for it to then interact. That would be another way to do it. That's also under the same enterprise terms of service that you're typically using. So our the I'm speaking specifically from the concept of anthropic. That's our that's been our tool tooling of choice. And we are we have ensured everybody's moved over. We've put some policies in place. We put some governance in place so that we're much more compliant with some of the opinions that have come out in recent history. So in order to do that for the person who's listening to this, like, oh my God, I have to get on a cloud enterprise. How does one do that? Yeah. Just go. Go. If you could probably just put it in here, your chat bot or Google, Google it and you can say account with Claude teams account, you're going to get a landing page. You can sign up, you can directly sign up for there. You don't need to work with a sales person in one of their orgs. If you want an enterprise account, then you probably will be having a conversation with them because there's some things that you'll you want to define together. And then if you want API back and forth because you're building custom applications, that you can also sign up just just online and enter an agreement that way. So it's very easy to do. A lot of people are opting to not to to not do it or to avoid it all together because they don't necessarily know what to do. But if you're using a free account, if you're not using a commercial grade account, you're likely putting yourself in a position where there's some risk. And I think the silent killer here is that if even if you do that and your team or your attorneys or their paralegals are using their free account on their phone or on their computer, you're still in trouble. So you have to make sure that not only do you sign up for a team's account, but your entire team is only using that account. Correct. You're going to want to make sure you you set that, you set that up and be you need to have some sort of the eye policy document in place and then a governance process in place so that you are proactively compliant. Otherwise, the the reality is, if you own a law firm, the liability rolls to you. It doesn't roll to your team member. So there are some edit, there's some protections that you can have as the owner. If you are doing the things that put you in the best position to make it a non-compliant decision that someone has made, nobody expects every single thing that's done within a law firm to be visibly seen exactly by the the law firm owner. But if there's no governance involved, then it's just you're kind of lacks a days of maybe putting a policy together. We all know that. We all have, you know. Employment employment agreements that are reviewed one time. We all have annual like our team member handbooks. Those are typically reviewed one time. Now in those instances, there's not a lot changing in the world of AI. There's a lot changing the utilities changing significantly. It's getting more advanced. And if you don't have a process for governance, you're kind of you're kind of just being negligent there and you're putting yourself at risk for not having a process for like, how are you reviewing how it's being used on an ongoing basis? And is your team staying compliant and at least doing something to make sure everybody's in, in using it the right way and leveraging it with the right tools and has very clear direction from you in terms of expectations on how it should, should and could be leveraged. Otherwise, you're just allowing the team member to fill fill the gap with whatever their preference, which which is not necessarily a good That's right. Well, I think what's interesting is the ABA has not issued anything since July of 2020 for their formal opinion. 512, which is the first national guidance on generative AI. Can you walk us through this a little bit? I think it's I think it's done because of the how it was written is more general architecture of how you should think about it, because obviously it's not it has nothing to do with any given state. It's just an overall guidance opinion. So and then different states have adopted different portions of they as they've written their own. So it's basically if you're if you're if you're in a state that doesn't have specific guidance, this is a good place to start. If you are in a state that has guidance, I would leverage this. And as a secondary add on to whatever the the state's guidance has been specifically. Yeah, that's that's your if you have if there is one use that one. A lot of them have come then coming out with different opinions. Some have just been adding to. The ABA guidance, some have been completely rewriting it and kind of putting their whole, their whole stamp of approval on everything. So it really depends. You have to do your own research on this. But there, there, there guidance was it hasn't been updated because there's it's a framework of how to think about this. And it's which is important because that's really what this is. This is a tool. It would be like, you know, any other tool that you use in your business, it's just a different type of tool that comes with the ability to do a lot more advanced things, rather than when I do x, y happens. This is when I ask for x. A million different things could potentially happen depending on how you've programed what you want the LM to do in between. So a little bit different. Yeah, I think our narrative to our attorneys is AI never replaces their judgment. Right. The attorney directs the work and they can leverage a tool to support it. But for anything specific to your situation, you got to rely on the attorneys that you have in their judgment. Yeah, we just did it. I was literally just in a training today about how to use it to draft a draft MSA. So we were we were building a skill as a team. And the whole the whole point of it is to like, how do we streamline process so that we can get into more of the strategic work versus. Just hunting and pecking for here's what the here's what goes here from this document and so on and so forth. You're still going to need to audit it, but it's going to it speeds the process up significantly. And it gets you to the more of the strategic strategic case work faster. And it your team in a significant way. So that's kind of how we how we think about using it. It's it's not a replacement at all. It is a it's a tool to kind of get to the meaningful work as quickly as possible, so that that can then be done with a higher degree of capability and with a more thorough process. Like we our our biggest use case probably is how we how we use it for actual trial prep. So like we we're going into. A hearing that's going to be a back and forth, back and forth hearing. We will set up our attorney, set up projects within Claude, and then it will create a lot of question sets and then evaluate all of those. And it does a really good job looking at the whole case data and then coming up with a lot of different unique angles. Now most of it is going to you're going to have to like like we've been saying comb through it, but it allows you to kind of repeat that process really fast and get way better, way better prepared for a trial than if you're doing it all manually. Because like you, it's going to generate different angles and different potential ideas and all of these things at a speed at, which is, you know, it's impossible for a human to keep up with it, but it's not going to be it's not going to be. It's going to be 100% accurate. It's not going to get everything right, but it's going to spur that creativity more and more. And then you get your first first set back. You know, maybe you're looking for, you know, 100 questions. You're going to get 3 or 300 or so from AI. You're going to sift through it. You're going to report back in the best hundred that you want, and then you can prompt it to give you more angles on this, this particular topic and that particular topic, so that you can really have a thorough, a thorough vetting of what, what you want to come into and what are all the questions that you might want to pose as you're, as you're litigating. So it's our team has found it super helpful for that. And it's it's been fun watching them embrace it and like leverage it for its actual intent because it's very powerful. It's a super powerful tool that it needs, you know, it needs a driver. And if you if you if you learn how to drive the tool, it can do awesome work for you, but you still need to do the the end result. You get lazy, you're going to get yourself in So good. Tony, thank you so much. Okay. So I want to go back to the building your AI governance policy. So we'll use what Tony just did is like the end of the episode. But let's cut back to how to Tony build his own or how did Sterling build their AI governance. So Tony, talk to us a little bit about the the governance policy that you put in place or help put in place at Sterling, because I think a lot of people are listening to this and they're like, okay, I got to get on a corporate account, which you have to do, and then I have to put a policy in place so that my attorneys can leverage it. How did you go about building that? And and what did the kind of final product look like? So we at Stirling, we first started from a research process. So Illinois has a really strong opinion on how we should be utilized. So that's kind of how we how we based. That's where we started. We used the ABA guidance as well. But we built out a policy. And basically what that spelled out is two types of accounts. Here's the ones that are approved to use. Here's what we've selected to use. And this is where you should be operating. So kind of put everybody on the team. If you're using AI here's where you're going to be using it. It kind of went through a whole bunch of different. Here's how we're going to use it. Here's how data is going to be used. Here's how it needs to fit into our client agreements. If a client ops out of wanting AI to be used on their case, that needs to be really clear. And like obviously then you know, you have your agreeing that you're not going to use it for that file. So there's a whole bunch of different things like that that we are then having the attorney sign as an agreement with us at the firm, and then we we're constantly reviewing that. So we're, we've we have an ongoing process of meeting with our teams and understanding of how, how they're using it, what they're using, so that we are essentially auditing their use cases and ensuring we are where, how, how it should be used is compliant with how the opinions have been written, just to keep our team safe and keep them honest. So all of their accounts are accessible to to us on the team so we can go in and do do these audits. And that's just something that we've we've taken on as kind of like an ongoing process. So it'll be something we continue to do. The benefit obviously, is we feel like we're putting ourselves in a position where if we I can't see it happening, but it could happen where someone gets lazy and decides that they're not going to really do the job the way they're supposed to. It puts us in a position where that's really going to go, that's going to hit them. Obviously, it'll be a bad black eye for the firm, but it'll be, you know, our firm owner's license will be much more defensively positioned than if we aren't doing any of this. So a lot of a lot of what we built from a process perspective was reviewed and confirmed with our two executive partners or to top attorneys, one from Wisconsin and one from Illinois. In terms of like how we wrote it up and how we interact with with it. And then those two are the ones that are kind of leading the warrior groups, as we're doing working group sessions on implementation of different tools and skill builds and whatnot. Perfect. So very simply, it's policy document. It's an ongoing process for observation and adherence. And right now there's it's I would say it's pretty manual, but it's definitely worth it because we've seen really, really cool production enhancements for our team that wouldn't be available through any other means. AI is here to help your firm. You just need to do it the right way. That's the second permit policy. Once you've got the account here, you know the specific rules as it relates to your state and an internal firm policy on AI. You are clear to make the next big decision, which AI platform should your firm actually build on? That's exactly what we're covering next week, so make sure to tune in. We're going to walk you through how to choose the right platform that works for you, and how to make sure you're building on the right foundation and not just the flashiest demo. make sure to follow the show. So next week's episode lands right in your feed. We will see you there.
.png)
