Subscribe
Share
Share
Embed
The Effective Lawyer teaches ambitious trial lawyers how to grow their skills and create a prosperous law firm. Using lessons learned by accomplished attorneys from around the country, we discuss lessons learned through their trials and tribulations. Our discussions cover a vast range of topics sought out by attorneys looking for advice, from depositions to how to market your law firm.
The show is hosted by Jack Zinda, Founder and Senior Trial Lawyer at Zinda Law Group. In less than 15 years, Jack and his team have grown Zinda Law Group from 3 attorneys to over 30, spanning several states and handling a variety of personal injury cases from gas explosions to truck accidents.
Jack and his guests share their knowledge and skills that they’ve acquired through the process of building one of the most successful plaintiff’s law firms in the country.
In each show we cover a new topic that an ambitious attorney would want to better understand, while providing practical skills to improve their legal practice.
For more information, visit https://www.zdfirm.com/the-effective-lawyer
00:04
Jack Zinda
Jack Zinda here today. I want to talk to you about one of those things that a lot of people are saying, a lot of things, but really nothing is actually being said. And of course I'm talking about artificial intelligence. It is on the news, it's in law review articles, it's in blog articles, YouTube, everywhere. It's changing everything. It's changing nothing. You're going to go broke if you don't do it. You're going to make millions if you use it. It's going to wipe out attorneys, paralegals, or it may do nothing. It can be very overwhelming.
00:31
Jack Zinda
And I thought it would be helpful if you just zoomed out for a second and talked about some practical tools and tactics you can do right now to help make you a better attorney, help your clients more, while at the same time a few things you want to never do. And I don't think AI will replace. First thing is take a deep breath. We are at the, you know, maybe one yard line. If this is a hundred yard dash of where AI is going, you know you're not behind. Very few law firms are using this successfully. In fact, a lot are probably spending a lot of time just trying to figure out what the heck it even is. So you're not behind and it's not going to kill your practice. So how can you use it though? How can you use it today?
01:07
Jack Zinda
So first you need to know a few. What are the tools out there that people are using right out of the box? Well, of course everyone's heard of OpenAI and everything I talk about today is going to be Based on using OpenAI as your platform to help you with your practice. Okay. OpenAI has a couple of different subscription models. You have the free model. Then you also have like a $20 a month and then you can get a business account that you can share with other team members. That can be around $200 and then it's I think around $20 for an extra team person. And then they have enterprise and levels above that. A couple things to know. If you do go and use OpenAI, you need to make sure you're using their HIPAA compliant version.
01:47
Jack Zinda
And also you have the right protocols to protect from sharing confidential information in the different tiers they have give you different sorts of protection. That's kind of the first buyer beware thing I'm going to talk about. Okay, so tactic number one, you can use AI to review your intakes, your intake team's calls or the transcripts. Okay, so how can you do that? First let's say you have no technology set up and all you do when an intake comes in is your intake team fills out a form and then gives it to someone to call the client back. Okay.
02:17
Jack Zinda
What you can do is you can take all the intakes that came in for that week, you can put those into ChatGPT and you can create what's called a custom agent to help come up with content ideas that could help your law firm put content on your website or do videos that could help you get additional clients in the future. So that's number one. You have to have no tech savviness for that. It's just the intake team filling out the form. You also take wherever that lead came from. If it was a Contact Us form, if it was a chat, if it was, someone told you about it. And then I can summarize that for you and you make that part of your weekly routine. Then you give that to your marketing team to have more content to create or you just create yourself.
02:52
Jack Zinda
Okay. The second thing you could do, you could have it grade how well your intake team's doing. Now, what our firm is currently doing is we have set up a software program where we will take the transcript from the intake call. We will put it through a custom built agent that then scores it puts out a score for the person that did the call and then attaches the score, feedback the transcript and how we should do things differently. And it gives it an objective score and a subjective score. On the objective score, what it's looking for, did they use the right terms? Did they say no fee, no recovery, did they know the right address, things like that. Okay, you don't have a software development team. How do you do that? How could you make that happen? Well, you could start simple. Let's see.
03:33
Jack Zinda
I want to make sure if my intake team is filling out all the information that I need. You could just simply set up an agent in OpenAI or ChatGPT. There's a place called Custom Agents. Or you can do a custom project. You could then upload examples of intakes that are filled out accurately and then you could train the agent by asking it, hey, I want you to review these intakes as I'm uploading and identify any areas where maybe something's missing. This sounds like it's really tax heavy and it could be difficult to do, but the way these tools are built, it's actually very easy. Now how can I help you with? Well, let's say you have one intake specialist who is not asking the right Questions. Maybe you're missing on source recovery, maybe you're missing cases with big damages.
04:11
Jack Zinda
And then you see this, you spot it in your report, and then all of a sudden you identify a case that's worth half a million dollars that you wouldn't have seen otherwise. The trick here is you want to try to delegate this to a team member to do this. And let's say you're really lost. Like, man, I cannot figure out how to do this. Use OpenAI to help walk you through it. Just go into the bot and say, hey, I'm trying to build this, please help me set this up. And then. Or you could delegate that to your paralegal or an assistant to help you do that. So that's number one. Use it to review intake team. Use it to review the leads, help you to score those, help you to put those together.
04:40
Jack Zinda
Okay, what's the next thing it could do to help you with your practice? Let's talk about another easy win. Okay, next easy win, depo transcripts. All right, you've taken great depositions, maybe you don't have a big team and you're getting ready for trial. You want to know, hey, what are all the instances that this, let's say it's a corporate representative mention this safety regulation. What are all the times that they talked about this incident that occurred? When did my client mention previous injury that they had? So you can build an agent that'll help you review your deposition summaries or your deposition transcripts to identify pieces of information you want. Now here's the key. It's really tricky with the agents, the way they currently work.
05:17
Jack Zinda
If you upload a lot of documents, the output you'll get will not be ideal because it misses a lot of things the larger the document is. So, so if you have a deposition transcript and it's really large, you want to make sure you're asking it in bite sized pieces. So practically speaking, what do you need to do? Well, again, you'd set up a project. We're all going to use these based on projects. That's what they use in OpenAI. And it's a little section where you say, I want to use OpenAI for this one type of use case. Now you're going to have to get the business account like I talked about it is definitely worth it. Even if you're on a shoestring budget, having this availability to use going to let you do things way more efficiently.
05:51
Jack Zinda
So.
05:51
Jack Zinda
So you set up the project, you give examples of what type of information you're trying to pull out of the deposition. So let's say I'm trying to pull all examples of where there's a previous mention of this truck driver having other crashes. So I would upload a deposition, then talk to the agent and say, this is what I'm looking for, this is what it looks like. And so then it learns what you're looking for. Then you would ask it to do the same thing on a new deposition, and you want to check if it's accurate. Once you realize it's accurate, you can start having it ask more questions to do the same thing. And then you could actually have that exported into either a PDF or answer that then your team members can put in a document.
06:28
Jack Zinda
If you put this into an sop, boom, you've got a great way to get deposition information right before trial, 30 days before trial, right before an expert designation or expert deposition popping up, and it can be on autopilot. These little things, like I said, they're not revolutionary where they're going to change your practice or you're not going to need five more attorneys, but they're going to add a lot of value to stuff you may not be doing or may not be doing well. And maybe that saves you 12 hours over the course of a quarter. That's still a lot of money. Next, let's talk about research. There's a really cool tool in OpenAI called Deep Research.
06:58
Jack Zinda
Let's say you're trying to find out something about an expert witness who pops up a lot of cases or a company, did they have other issues involving gas explosions or a terrible driving history. We have a really crazy case right now involving a cemetery, and we're trying to find other things involving cemeteries and things like that. So in OpenAI, you can go and select a button that says Deep Research and say, hey, I want you to look up everything involving Smith's gas company and gas explosions about them that have been in the news in the southwest United States in the last three years. And what's cool is it's going to take its time and actually do as much deep research as it can across the Internet and across other resources it has. And it's going to take maybe 15, 20 minutes.
07:37
Jack Zinda
And you actually see what it's looking for as it's doing this research, and then it's going to come back with a lot of stuff that you can use that hopefully has good information. You also want to make sure you ask it to cite whatever it finds.
07:48
Jack Zinda
So.
07:48
Jack Zinda
So you can then go get those resource materials to Find out if it's helpful. Now, how can this help? We had a case recently. We were trying to figure out where venue would be and there was one venue location that would have been really bad for us and tough for the client and one venue location that we thought would be really strong. We weren't sure. We were dealing with a pretty large company if they had a nexus there. And sure enough, with Deep Research, it found a really great nexus or an advertisement this company was doing about doing business in this exact county. We had not found that it was almost like a small marketplace advertising platform that we hadn't thought of.
08:16
Jack Zinda
And that took 15 minutes and that probably made our firm lease extra million dollars on the venue of that versus the other one because of the location, the jury pool, all those sorts of things. And that's just one example of what you can use using Deep Research. But again, that's a really cool feature that if you're not using, check it out again with all of this stuff. Buyer beware. You need to be careful because you are a lawyer. You need to be a lawyer. Do not trust the citations. If you do not check it yourself. You don't want to go to court, make an argument and then actually be factually wrong. This is where I don't think it's as revolutionary as everyone says. Garbage in, garbage out.
08:50
Jack Zinda
It is using this, the Internet, to come up with its answers, which can be great, but it can also be terrible. That's why a lot of the use cases where it's actually doing something that's repeatable, that you trained it, actually it's a little better than others. Next thing you can use medical records, bills. So a lot of us, we either create medical timelines or hire companies to do medical timelines. But sometimes I'm looking for specific pieces of information that maybe is not in the timeline. So for example, if I want to find every time they took a painkiller in both previous and current records, I can have OpenAI go in and search pages at a time to try to identify that.
09:23
Jack Zinda
Now what's key here is you want to make sure when you upload the documents that they are OCR and that they're legible because otherwise it's going to miss some of it and no, it's not going to be perfect. So let's say I want to find every instance that my client took oxycontin somewhere the defense is going to make an argument that they had drug seeking behavior, they're malingering or something. So I'm going To look at the records, see, okay, are these legible? Are there parts that I can't read because they're maybe handwritten? I want to take that into consideration. And then I'm going to make sure the records are OCR'd. Again. I'm gonna set up a separate project in OpenAI. And then I'll say, hey, review pages 1 through 12 for any mention of OxyContin or similar substances. Then it's going to review it.
10:02
Jack Zinda
And then I would double check, is it accurately giving that information? And then you want to have it go to the next pages and the next pages. And then at the end, you want to have it summarize it and pull it all together along with citations. And this is something you can very easily train your team to do. And if you get a really tech savvy on this stuff, you'd actually a simple protocol where it'll do that automatically review those small chunks at a time. But that can be a huge time saver right there. That alone. Let's talk about another thing you can do here. Generating content. We hit on this a little bit before. Now, again, you don't want to just use this as a tool to avoid actually putting out good content.
10:37
Jack Zinda
And I'm talking about content that you can use to both market to old clients, market to new clients, as well as things that are going to help make your firm look good in the community. So let's talk about marketing for cases you want. The first thing to use OpenAI for is like, hey, come up with the different cases that my firm handles based on our website and actually surf your website for you to pull those out. Okay, Then maybe upload a bunch of cases you resolved or settlement statements or demand letters of cases that you resolved. Hey, come up with different case types based on these cases that we got big results on that we could put marketing material on. So then it comes up, let's say another 15 to 20 different case types or ideas you have.
11:12
Jack Zinda
Okay, now let's come up with five to 10 questions that someone might have about these topics. Now let's come up with five to 10 things they may need to do to help with that topic. All right, now we have an outline. Now we have information we can put together. And now what I'm probably gonna do is maybe do an audio where I go through these topics in an audio format, send that into ChatGPT, have it transcribed, say, okay, now take my audio, take the outline, put together and put together some written content for me that I can now put on the Internet that I can use to get other clients. You could also ask it to put in the SEO information, the meta tags, metadata, all of this great stuff that it can help you with.
11:43
Jack Zinda
Again, one thing to be careful is if you're going to use it for marketing, make sure you are having to rely on things that are best practices. So you might find a book, for example, if it's SEO, say, hey, rely on SEOmoz when making these decisions on the SEO data to include in that. That may be second level, but if you're looking to this stuff yourself, it's possible if you take those approaches. Next up, let's talk about client communication and draft documents and things like that. The first thing I would do is come up with how you're going to communicate with people you talk to all the time and use OpenAI to help draft letters for you email communications, as well as summarize things that occurred in a case.
12:21
Jack Zinda
For example, let's say I have a similar email I send to all of opposing counsel. I may train ChatGPT, which is the software program of OpenAI, to draft emails a certain way. If it's from an attorney or what I'm sending to an attorney, I can upload the emails that I would use that I typically use and I'll say, hey, please come up with an outline for an email from me to an opposing council based on the ones I uploaded. It'll then upload an outline.
12:44
Jack Zinda
You can ask and put together a template, you can tweak it and then every time you have a situation like that where you're emailing opposing counsel, you can do a voice memo with generally what you want to say, have that uploaded into the agent, and then you can have a bunch of emails for the day drafted for you, or letters, good to go. Just have your paralegal or assistant pull those out, put them on your letterhead print. Probably saved you eight to ten hours right there. And it'll learn how you like to talk to people. The same with client communication. You can have your team have a shared project in OpenAI that just has ways we communicate to clients, whether it's emails, whether it's letters, whether it's settlement statements.
13:17
Jack Zinda
And I'm not talking about the actual drafting the document itself, but how do we explain it to people and how do we include it in their correspondence? And you can have your team share that in. Boom. You have a really clear way of putting together documents that everyone's going to be and find useful. Now with all the stuff we just talked about maybe we came up with 100 hours of savings in a quarter. Between your rate, your paralegal rate, your legal assistance rate, that is a big savings. Plus you probably made a lot of money by finding either a different venue to go after, okay, maybe come with new case. That may be the next million dollar case you got in the way. You approach this all without having to hire an expensive agency or add more people to the team.
13:54
Jack Zinda
A few parting thoughts on this. One thing you should never do is let AI be a lawyer for you or a paralegal or have it send correspondence that you don't review first. You will look foolish and you will put your bar card on the line. It is still nowhere near where it can be an attorney. And this is where, when I think a lot of lawyers that are worried about it putting them out of business, I think it will put lawyers out of business that are not adding much value. If your practice is essentially, you know, hot docs on steroids, yeah, AI is going to put you out of business. But our job of lawyers is to give advice, to pay attention to details, get goals accomplished. And there's always going to be a place for that.
14:31
Jack Zinda
And so as you're using these tools, you need to make sure that your bar cards on the line, you're reviewing everything before it comes out and assume it's going to make errors. Okay? Assume that there's going to be mistakes you have to find and then tweak it and come in with those expectations with a lot of things. If you come in expecting to be perfect or expected to do your job for you're going to be sadly disappointed. And I would hate for anybody to get a bar complaint or commit my practice, say, well, I was following the advice I heard from Jack. So my last piece of advice is be an attorney, think and use these tools. They're going to help you a lot. Till next time.