Executive Summary

In this episode of the *Grow Your Law Firm Podcast*, Ken talks with Aron about how his disability law firm is using AI in practical, in-house ways. Aron explains how tools like Claude Code are helping his team build lead generation tools, custom dashboards, and case analysis workflows without hiring outside developers for every project. He also shares how the firm is using AI to handle major pain points in disability practice, especially large federal court transcripts and statement-of-facts drafting. The conversation also covers data privacy, how to introduce AI to staff without creating fear, and why law firm owners should start using these tools themselves to understand both their power and their limits.

Introduction

Welcome to the *Grow Your Law Firm Podcast*, brought to you by PILMMA.

This podcast helps lead lawyers to more growth, profit, and freedom.

Here is your legal marketing expert and host, Ken Hardison.

Ken: Well, hello everyone, and welcome to another episode. Aron, do you have 10 minutes to share what he is doing with AI in his law firm? Aron, welcome to the show.

Aron: Well, thanks for having me, Ken. Excited to be here.

Ken: Yeah. So tell our viewers and our listeners a little bit about you, your journey, and then we will get into the amount of AI you're doing yourself. Give us a grounding.

Aron: Yeah, sure. Well, I'm a partner at Hiller Comerford, a disability firm. We're based out of Buffalo, New York, and we've got over 240 employees, so we're a pretty big operation.

Mainly a disability practice. We've got over 18,000 Social Security clients and over 2,000 VA clients, so we do do a significant amount of volume there. And we've got a small but growing PI practice that we're trying to add to that.

Ken: Absolutely. Good deal, man. How long have you been doing disability and doing this?

Aron: I've been doing disability since 2012, when I got out of law school. This firm, Hiller Comerford, formally known as the Law Offices of Kenneth Hiller, my father started the firm back in 1999. So disability law has kind of been in my blood. I remember being like a 10-year-old kid driving in the car, and he would have the tapes from the hearings. He'd be listening to them, the disability hearings, and we would be talking about the cases and stuff.

Using AI to Build In-House Tools

Ken: We were talking at the National Summit, I can't remember where, about what you're doing with AI. I think a lot of lawyers out there really are scared of it, or don't understand how to use it or what the use case is. So you graciously said it would be good to talk about what you're doing with AI in your law firm, how it's helping, and what are some things people need to be thinking. What are you doing? How are you using it now?

Aron: Yeah, I mean, I think the last six months or so have just been the final nail in the coffin for anyone who thinks this is just hype and it's not going to really impact things. It's been a wild six months.

Really starting, I think, for us, with us becoming aware of Claude Code and the stability now for people who, like myself, don't know how to code. I can basically code somewhat sophisticated apps and projects just by talking with Claude. And that's a game changer if you're looking for things like deploying your own Facebook lead generation campaigns. We don't have to pay someone to do that anymore. In a few hours, we can set it up ourselves.

So for VA, we've got a disability calculator out there on Facebook that says, you know, find out how much money you could be making, and then people click on it. And then at the end, it asks for the contact information. We've been getting a really good cost per case on that, and that's something that we made in-house. We didn't have to hire anybody, and it looks great. The disability calculator looks great.

And so that's just one thing. Custom dashboards. We've got our own CRM, SmartAdvocate, and it comes with a lot of reports. But, like I said, we're a disability firm, so a lot of the reports in a lot of these CRMs are geared toward PI. So we don't have to go out and pay somebody to create dashboards for us. It's, again, a couple of hours, and we can have any dashboard we want.

We've also got a tool to analyze our med-mal cases. That's been a big pain point for us because it's kind of hard to come up with hard-and-fast rules about med-mal, so it kind of requires an attorney to review it. But we're still not going to take 99% of the cases. So we've got a med-mal analyzer, and over time I think it's going to get better for us.

Privacy, Data Security, and Enterprise Agreements

Ken: So with Claude, can you sign an enterprise agreement and keep it private? Because a lot of lawyers are worried about client stuff getting out into the upper world.

Aron: Yeah. So, for all the things that I talked about, we're not putting client data into it. But yeah, we do have an enterprise agreement. So if we did want to put client data into it, I think we could.

Ken: Yeah, I think that's the key. I know Jim and I understand this. With ChatGPT, we would not sign in through a browser. We could pick a private-file agreement where they're not training on it. Right?

Aron: That's right. That's right.

Ken: Because that seems to be the big deal everybody's worried about more than anything else. And then there are so many companies out there talking about this, this, this, and a lot of them, instead of doing what you're doing and doing it themselves, they want you to broadcast it.

Aron: Right.

Why AI Makes In-House Software More Realistic

Aron: One of the things we're actually considering doing is we've got work in a center and hiring our own internal software engineer because, I mean, I'm able to create these, what I described, which are incredible to me that I'm able to create these projects. It's kind of cool, but they're pretty simple.

If you get somebody who actually knows how to code and now you get them with Claude, they're able to do things in six months that previously would have taken a lot longer. So we're kind of exploring that. I'm seeing if we can recreate some of the software that we pay for. If you have an in-house employee and it takes them three months to recreate it for you, that could be a win.

AI for Disability Cases and Massive Federal Court Records

Ken: As you grow bigger, especially in disability, you've got so many records. I've never practiced in it, but there's a lot of stuff to go through. Are you using AI to sift through this stuff and get some judgment on it, whether it works or not?

Aron: Yeah, so for Social Security, actually, it's really helpful because one of the things that we do, we have a very active federal court practice. What happens in federal court when you file a Social Security appeal in federal court, they submit a transcript of everything that happened at the agency.

And these transcripts are 1,000 to 3,000 pages. We were working on a 12,000-page transcript the other day. You've got to take that and put that into a statement of facts as part of your brief in federal court. So that is a huge pain point for our attorneys, to say the least.

So yeah, we've signed on with Eve, and we're really happy with the statement-of-facts output that we're getting so far. I mean, we haven't been doing it too long, but I've been going through them, and everything is accurate. And I'm just amazed that you're able to get a 15, 16, 20-page statement of facts written at a level that you could submit to a federal court. That's what it's producing. I mean, it really blows me away.

Ken: Of course, you've got to look over what it spits out, though.

Aron: Of course, yeah. And we have a policy here about that. We promulgated a policy that you're never allowed to submit anything that's been generated by AI unless you've independently verified the accuracy of it. That's something we talk about a lot.

Frankly, our attorneys, we don't really have to remind them too much because they're nervous about it anyway. And it seems like there are plenty of attorneys in the news who are abusing AI, and that kind of keeps it fresh in our mind that, hey, we've got to check this stuff.

Ken: Yeah, sanctions have been some big things, and they have been in the last month.

Aron: Yeah. A lot of the people who are being sanctioned, not only did they use AI in kind of a reckless way, but then when they get caught with it, they don't just fall on their sword and say, sorry. They fight back or something. Maybe not the wisest litigation strategy.

The Future of CRMs and Case Management

Ken: So what do you see in the future? What's on your plans? I mean, you've done what you've already done. What do you promise for, as you said?

Aron: Yeah. So the way that I see things going is I'm kind of wondering if, and the CRM companies probably won't like to hear this, but I'm sort of wondering if AI is really going to totally displace the way we think about case management, where all of your information could just be in the AI system, and then you're interfacing with the chatbot to get any information you need out of the system about the case, about whatever you're working on.

Instead of having kind of a rigid interface, you could see that there could be some benefits to that. And I know some of the AI companies themselves want to move in that direction as well.

Ken: Oh, it will be interesting. I think for us it's really important. We're trying not to sign any long-term agreements for anything these days because everything is changing so fast. So anything with technology, we want to keep our agreements certainly under a year, but preferable is just month to month because who knows. Something could come out next month that just totally changes the game.

We were just discussing that again earlier today, and that's the big concept. It's changing so fast.

Aron: That's something we learned early on. Right at the beginning of this whole AI trend, we signed on with a company and we did get locked in on a longer-term deal, and we really regret it. So we don't use that tool anymore, and it was a hard lesson. It's certainly a mistake we won't repeat again, for sure.

AI, Productivity, and Fear of Job Loss

Ken: Good deal. So show me about the tools that's making a difference. Are people using it? Are the people that you are using it with being more productive? Where do you see it in terms of the whole? It's not being left to tell.

Aron: Well, it's hard to tell because we're trying to grow, and so we are growing. This is what I tell my employees. I'm not looking here to improve productivity and get rid of employees and just kind of rest on my laurels here. We're looking to grow.

So I think the way we're using it and the way we're thinking and talking about it is, we want to grow. We want to compete with these huge mega firms. So this is going to help us do that in a cost-effective manner. We may be hiring slower than we normally would as we grow, but we really haven't sat down and said, geez, now we can do this so much faster, we can get rid of X, Y, and Z.

If anything, something that has happened is we've applied, let's take Foundation AI, which helps you sort your incoming documents and get them filed in your CRM. Well, that's really lowered the demands on our mail room. We don't, we're not getting rid of those people. We're just finding something else for them to do.

And so, yeah, I think that's a really important part for law firm owners because if you don't even have to portray it this way, people are going to be nervous about the AI taking their job. So anytime we talk about it, we like to talk about it in kind of an abundance mindset, like this is going to help us do more.

I gave the example of the federal court brief. It's helping us do the statement of facts. Well, in federal court you get paid for your time. So we don't actually want them to take less time on their cases because we would make less money. What we've really presented to them is, look, this is a way to eliminate a pain point for yourself and produce top-quality work that's going to really increase your win rate and, in that sense, make more money for the firm.

I guess what we really try to talk about is, this is just going to help us do more work. When you think about the legal industry as a whole, I'm kind of thinking lately that, I think it will be disruptive, there's no doubt. But when you talk about these predictions of mass layoffs and stuff, that assumes that there's a cap on the amount of legal work that can be done. We can always do more legal work. We can always do more. There's more clients we can serve. There's more cases we can take on. So I just think it's premature to assume that there's going to be all these people out of a job.

How to Roll AI Out to Staff

Ken: So how do you go about communicating what you just said? You're saying you could call them, nominate, talk to a manager, do something on email maybe. How do you do that? I think if I was running a firm, I would be wondering how that's the best way to communicate that statement. And your firm's big, so it'd be hard to do it well.

Aron: Yeah. I guess all of the above. We do it all. We try to just constantly communicate about these tools and that they're important to us and that they're not here to replace people.

I like it when we can have an ambassador, an employee who understands the message we're trying to get across and they can deliver the message because they understand what the employees are going through more than I do. So, for example, we had one of our top trial attorneys talk about Eve to the PI staff before we rolled it out, and I think he gave a better presentation than I could, simply because he was more in tune with how the employees are thinking about it.

One of the things he mentioned was, look, in personal injury especially, there's a personal touch here that can never be replaced. I thought that was a very effective message to the staff. I saw a lot of people nodding.

So we do small groups. We do a larger, like when we rolled out Eve, it was a large group. But then we follow that up with smaller groups. And then we try to meet individually with people who are resistant and see if we can't do a one-on-one and get them on board.

Ken: Yeah. Have you had to fire anybody because they wouldn't use it?

Aron: Yeah, we haven't gotten there yet. We haven't gotten there yet.

Ken: I've heard of it happening in other businesses. If they don't use it, they've got to go.

Aron: Yeah, I mean, I've heard stories of that too. I think the way that I prefer to do it is to say, look, if now because of the AI the productivity is increased to a certain level, we expect you to meet that level regardless of whether you want to use the AI. That's fine. You've got to meet that level that everyone else is performing at. That's kind of the way that I like to think about it.

And so far we haven't had to get rid of anybody. But, like I said, we've really just exploded in our adoption of AI over the last six months, so that could be coming.

Why Law Firm Owners Should Start Using AI Themselves

Ken: Do you think there's anything I have not asked you that you think is important for all your peers out there who are struggling or trying to figure out what it's for?

Aron: Yeah. I just think you need to start using it, being a consumer of it. You just got to start playing around with it. I really think every law firm owner should just play around with Claude Code just so you can understand what it can do.

Regardless of whether you want to do this yourself or not, I personally like doing it, but regardless of whether you want to do that, I think it's important to play around with it, and it's important that you have an understanding of these systems, how they work, what their limitations are. If you don't have it, you're not going to be able to really explain it to people that you want to use it.

Ken: I agree. That's the way I started out. I started playing with it. I had my trips trying to figure out certain things, stuff about where's the misplacing of the, you know, right? It's a piece of it. Something like that. Or it's not thinking about going nearly, or like the 10 things. But me, I can do further 30 albums. You understand.

Aron: Yeah. I mean, it's great for planning. If you need to plan a project, it's great for that as well. But yeah, by playing around with it, you get to know its limitations, what it can do, what it can't do, and how to get it to kind of give you what you want. Because there definitely is a skill to that, and you'll never get it until you start doing it yourself.

AI for Brainstorming and Forecasting

Ken: Yeah, I read a book called *AI Leadership*. I don't know if you read it.

Aron: I have not.

Ken: He was talking about, probably adopt the way you would think about it. He uses it to solve problems of businesses. His big deal is you throw in the context, tell the story of what's going on and what the problem is. And then what he would tell ChatGPT, Claude, whatever, he says, now ask me one question at a time if you need anything, I think three to four. And then just do it and ask your question, and you can answer them. And then it says, I've got enough information. This is one way to handle it. This is another way. And then you can talk through, what if this wasn't there?

I've used it before, and I can tell you that I wanted five different ways to approach something, and three of them are good. Two of them are great. One of them I had never thought of. It was confirming to me. I didn't even know. I just never saw it. It doesn't, I don't know, it's like I forget. All that was in a reach.

Aron: Yeah. Great for brainstorming. And we use it for something kind of similar. We've used it to kind of help us do some advanced financial forecasts, and that maybe I can understand, but I wouldn't be able to do myself.

So you can give some of your KPIs, and then you can have it run pretty sophisticated statistical analyses. For example, one of the things that I would never be able to run myself or figure out how to run myself, it can project my cash flow based on a certain number of KPIs. And I didn't even determine what the KPIs were. I asked it what the KPIs should be.

So we built this kind of from the ground up, and then you put it in and it runs an endless number of simulations. Then it gives you kind of like, here's your worst-case scenario, bottom 25% scenario for cash flow over the next couple months. Here's your top-quarter picture, and here's the middle.

I mean, that's a useful tool for me, for sure, and I would never be able to do that myself or even know how to set it up. But just through talking with Claude Code, it's not hard. Anybody could do it.

Ease of Use and the Human Analogy

Ken: Yeah, I can do it. I mean, I just talk to it. I say, you know, I'm not out in the world for the tech side like anybody in the world, but I still use AI every day in my life now, and it makes my life easier.

I'm worried about some things down the road, it's getting out of hand, but I think Congress and all will have to deal with that. But for me right now, personally, it's kind of big, perhaps so. If you're out there listening to this, at least, and you're 67 years old, you don't have a lot of time getting there and using it and getting somebody to do it, but anybody can get it. That's my take on it.

Aron: I think it's very user-friendly. I mean, if you know how to ask a question, you know how to use AI.

Ken: Yeah, and we're all here. We're supposed to know how to ask questions. That's where it's from.

Aron: That's right. That's right.

Ken: That's what they train us for a little bit. Now that I think about it, it is sort of like a deposition. Your hair helps.

Aron: Yes.

Ken: It asks me a question, and then it gives you something, or like, I don't know, that's not really what I mean. Like, how about this? So it is similar.

Aron: It is. It is.

Why AI Fits Disability Law So Well

Ken: So it's great. And then I'm actually very proud of what you're doing. People don't realize how that scales. The 18,000, I've had maybe 3,000, and I thought I was big. So I can imagine six times that. I don't know what you would do.

Aron: Yeah, I feel like if you have a disability practice, because it is heavy, it's very repetitive, this really can help you. I think more labor-intensive, I don't know what to do about that.

I mean, I think it's just tailor-made for our practice groups, and that's why we're leaning into it so hard. Especially if we want to compete with, you know, now the private equity money is moving into the disability world as well. We gotta be able to compete with them.

Ken: Yeah, absolutely. That's a good deal.

Closing

Ken: So are you going to Summit this year?

Aron: I'm sending my marketing director. I did so much traveling. I'm not sure what time zone I'm in at the moment, but I am sending my marketing director there. It'll be his first PILMMA Summit, so I'm excited for that.

Ken: He'll probably come back overwhelmed.

Aron: Yeah, drinking from a fire hose for sure.

Ken: Absolutely. Okay, my friend. Any parting words before we take a holiday?

Aron: No, this is great. Thanks for having me on.

Ken: Thank you. I appreciate it, because I know in that committee case, you've got to be big.

Aron: Yeah, that's right. That's right.

Until next time, this is Ken Hardison, dedicated to your success.

You've been listening to the *Grow Your Law Firm Podcast*, the podcast that leads lawyers to more growth, profit, and freedom.

Go to GrowYourLawFirm.com to find more ways to market and manage your law firm.

Please leave us a rating and review wherever you listen to your podcast.

⚠️ 🛠️ `run for f → run do → print lines 1-220 from $f → run done (agent)` failed