New Trellis AI Solution - Leverages State Court Data to Automate Litigation Tasks - podcast episode cover

New Trellis AI Solution - Leverages State Court Data to Automate Litigation Tasks

Dec 11, 202417 minSeason 1Ep. 13
--:--
--:--
Download Metacast podcast app
Listen to this episode in Metacast mobile app
Don't just listen to podcasts. Learn from them with transcripts, summaries, and chapters for every episode. Skim, search, and bookmark insights. Learn more

Episode description

LawSites' Bob Ambrogi speaks with Nicole Clark, founder of Trellis, about their new Trellis AI tools.

Trellis AI is a new legal productivity platform that leverages the largest repository of state court docket data to help litigators evaluate cases, automate brief drafting, suggest winning strategies, and more.

To read more about the news, head over to LawSites Blog, or visit Trellis.law

Transcript

How Trellis' data set differ from other tools

Welcome to LawNext PR, the podcast where we put a spotlight on the latest news coming out of the legal tech industry. This is Bob Ambrogi, and in each sponsored episode of LawNext PR, I interview a legal tech company about its just released news or latest developments. Today we're talking about Trellis AI, a new set of generative AI tools from the legal technology company Trellis designed to simplify litigation tasks.

Joining me to discuss the new feature is Nicole Clark, the founder and CEO of Trellis. Nicole, welcome to the show. Thanks so much, Bob. Great to be here. Always good to see you. So, before we get into talking about Trellis AI, that which you recently launched. Why don't you give us a little bit of an overview of Trellis and what you do?

What is Trellis

Absolutely. So Trellis is a state trial court research, analytics and AI platform. And what I mean by that is for anyone who has worked at the trial court level, ever been in state court, they know that the state trial court system is a massively fragmented and opaque system. And what we do is we go in county by county. We aggregate the state trial court data, we structure it, normalize it, make it searchable for really a single source of truth for the United States, trial court system.

for the United States, trial court system. And then, of course, we're able to then power analytics like judge analytics, law firm analytics, and now AI. On top of that. You were yourself a litigator before starting this company. So you you know, the paying a lot of lawyers, felt right. I know it well, I was an employment litigator, which means I was in state trial court all the time. And I simply couldn't believe that, a product like Trellis didn't exist yet. How did we.

How did we have, you know, such a great ability to look into court of appeals and to look into Supreme Court. But yet in the court system where every case starts trial court, we just had a black box. So you've just recently launched Trellis AI, which is a really kind of a whole new set of features and functionality, on top of what you had already built. So why don't you tell us about Trellis AI on a high level, and then we can dig into it a little bit more? Absolutely.

So, you know, with with the data So, you know, with with the data that we have, we have this just absolutely incredible data set at a massive scale and, a way that it just is unmatched. And because we have that, we have just hundreds of millions of motions and briefs and really what is effectively attorney work product. And we're able to use that to help litigators. We really work for the product is for trial court litigators.

And we're able to help them do litigation tasks that they, would otherwise be doing more manually. And we're able to do that because we have this incredible data set of which to train and to, uses as examples. to, uses as examples. So, can you talk a little bit more about the data side, just in terms of how extensive it is and how that, how it perhaps is different from what other companies out there in the market might have? Absolutely.

So attorneys have traditionally used really Court of Appeals research, which of course makes sense. Court of Appeals is binding on the lower courts, and the lower courts are supposed to listen. The truth of the matter is, at the trial court level, judges really do anything. And and anyone who has practiced in state trial court knows quite how unpredictable that court system is. And part of its unpredictability is the the, impossibility of trying to understand what happens in that court system.

And so at this point, we have over 2600 counties. Again, every county is a fragmented, unique universe onto its own. And then on top of that, we have hundreds of millions of motions and briefs and judicial rulings.

And then from there, we're able to synthesize information out, like practical information on how what the particular judge that you're appearing before, how they rule on similar cases, what their motion grain rates are to how a particular law firm, really, what does their performance look like across the nation? So we're now at 45 states, 2600 counties, which is over 3000 courts. And this is the data set that we have.

Just think of it as the entire United States court system, which is so much larger than simply court of Appeals, because every case starts in trial court and 99.7% of cases don't make it to trial. They live in that state trial court system, and they die there. And it's rare, actually, to make it all the way to trial and then to court of appeals, which is the subset of data that was historically used for research. So how does all that data help power these new AI tools? You've launched?

So think about it as some of the work that we have watched our customers and our litigators do more manually on our system. So let's take motion drafting for example. We've always been heavily utilized for motion drafting because we have hundreds of millions of examples of motions on before a particular judge on a particular cause of action with very similar case facts.

But historically, lawyers were pulling those more manually and working with that to create sort of to use as an outline and create a motion. What they can do now on Trellis is by practice area, by particular cause of action. They can draft msj arguments in the way that we do that is we have the complaint, because of course we have everything that's filed in the we have the complaint, because of course we have everything that's filed in the in the state trial court system.

We extract out all the important facts. We search across our entire database of similar cases to find cases with the most similar facts. We then look at msj arguments in those cases and arguments that have actually won. We take argument and case law that's able to use to generate the argument. And then we take the facts of the attorney's particular case.

So they're working on and apply the facts of their case to this winning argument in law, and provide them with the output of these msj arguments, which are absolutely, incredible. And, and work to create such a, such a strong starting point. And it's not just sort of random Court of Appeals text generated, but it is able to use the hundreds of millions of documents that attorney to filed and where they've actually won.

And it's able to use that case law and those arguments to create this draft output for the litigators. So the AI is getting trained against the arguments that have worked, that have won the cases that have worked then and won in other cases, and applying it to your factual situation. We said it even better than I did. Well, I'll just try to summarize, And you mentioned motions for summary judgment. It's also motions to dismiss as well. Right. That's correct.

And we're obviously going to continue building on from there. Think of this as a starting point with quite a quite a bit to tackle, Think of this as a starting point with quite a quite a bit to tackle, but we're trying to go after the most common or most difficult, most time consuming motions as a starting place. Right. And, and so something else that, there's these AI tools let, a litigator do, is to create a kind of an assessment of, of the case that you have at hand.

Right. Can you explain that? Absolutely. And and this is, one of the a features that our customers are really excited about. So what we do here is we allow from just the start of a case where you have the complaint, and what we do is we use all of our data, similar cases, similar outcomes. What verdicts look like, what juries look like. And what we do is a full case assessment based on that. So what are some of the tasks that attorneys need to do right away?

Creating timelines, creating chronologies. You know, identifying witnesses to depose, identifying witnesses that you need to find and may have valuable information that you'll need to notice depositions. So a lot of really practical, tactical information, So a lot of really practical, tactical information, but then mixed with a high level about the case, what are the legal issues in this case? What are the elements that need to be proven from a very start?

How is plaintiff attempting to prove those from as a starting place? How will they need to continue in order to win? What are common defenses that have been really successful against these arguments in cases with similar facts? So think of it as the very high level. What does a jury pool look like? What are recent verdicts look like for similar cases, so that you can see the case for a big picture from the very start? What do I need to prove and what have other cases?

What has developed in those cases? It could be helpful for me from a starting place then down to the tactical, given these facts, given these causes of action, what are motions that you should consider? What are the next steps that you need to do? What should you talk to the client about? So we're trying to really give a a of bull case assessment, evaluation of a case from the very start, which can be used in so many different ways.

One of which is, obviously, when you're just starting a case to be able to pull it all together, or when you want to communicate that to the client and you want to pull some of this information together for the client, or when you're having new folks, new colleagues come and join in, they need to ramp up quickly on the case and understand everything.

But it's also helpful for others when they're thinking about evaluating a case in general for, you know, whether to take certain cases or what are going to be, you know, the, the obstacles in a particular case and knowing those quickly upfront. And so we really try and make the this incredible thorough evaluate option that gives both strategic and tactical information based on similar cases. Yeah. So really becomes like a playbook for the case. And I this is this is a podcast.

But I would encourage listeners or watchers even to go to your website, go to I wrote a blog post about this. It's really it's really quite cool. I have to say. You said, you said that your customers find this is one of the most interesting or one of the most intriguing features. I also thought it was really, really cool. And, it's it's a highly detailed, outline of the case.

And then, as I understand it, as the case progresses, you can continue to update this with additional documents and additional information and continue to refine the the entire case assessment as you go forward. Right? That's correct. Absolutely. And I would say for that, one of the things that we've worked with some of our customers, you know, we how do we get attorneys the right information at the right time.

And what we're doing with some of our law firms is creating workflows where they get the attorneys. Sometimes with the dinger system, it doesn't matter which dinger system you use or which court alerting service you use, but being able to have the case assessment right there with the new case that you're identifying or giving the attorneys the this information directly in their inbox when the case gets assigned to them, without them even having to go on to Trellis and to request it.

But we work with firms to do that and create those automated workflows as well. So we've been talking about these kind of case level tools. In addition, this roll out of Trellis AI includes documents specific or I think, as you've called them, on document tools when we've talked about it before. Can you describe what those kinds of tools are? Absolutely.

So if you think about the vast database that Trellis has, we obviously have these hundreds of millions of motions and complaints and filings and, we are able to do a variety of tasks on all of our documents. So if you need to create a timeline, if you need to create a timeline off of multiple documents, all part of a case deposition in information and, what have you create timelines from there.

If you need obviously to summarize or extract information to extract the case law that's been cited to you so that you can then research that, there's so many different, tools that you can use. there's so many different, tools that you can use. You can, analyze arguments. You can, analyze arguments. What are the main arguments that they're making? What are they using to support that? What are the facts?

They're supporting that with all the way to, what are, affirmative They're supporting that with all the way to, what are, affirmative and OAD offenses that could be used based on, what we've seen in the, millions of other, cases that have had similar issues. millions of other, cases that have had similar issues. So there are a lot of different document tools like, but those are those are some of the highlights. Yeah. But I mean, it's again, just to, go a little deeper into that.

I mean, you can take a complaint tool, you can take a complaint and analyze the complaint, and it's going to tell you, not just the causes of action. Now, it's not just what's in the complaint, but what's needed to kind of prove the allegations of the complainant and what facts may be, what facts are alleged in the complaint, and maybe strengths and weaknesses of the facts as they're alleged. So it really goes beyond simply this is not just a summary.

In other words, this is a real kind of analysis of a document. Is that fair to say? That's absolutely true. Yeah. Yeah. And and the, the, the documents that you can apply these tools to really can be any document. They don't have to be something that was already in on Trellis that was came out of a docket on trolleys. If a lawyer wants to upload a document or something else, they can do that as well, right? Yes, absolutely.

So we obviously because of our database, have, hundreds of millions of documents, you may not even need to upload yours. But the good news is that you can upload any document and we can perform the same tools that we do on a document that you've uploaded. So really trying to make this great if the data is already on Trellis, but if it's not, don't worry. All of the our same, document tools can work on any document that you upload.

Right. Now, I believe it's still the case when we talked last, about this product, this was available for California cases. The case assessment tools are available for California at this point. Right? And then will be expanded? Case assessment is actually available across the nation for okay okay. The motion drafting summary judgment motion drafting is currently California okay. And you plan to expand that at some point into other. Absolutely. We're we're actively working on expanding right now.

Yeah. What else would you like listeners to know about Trellis? I that we haven't talked about. What I, what I would say is one of the things obviously, our data What I, what I would say is one of the things obviously, our data and or the foundation really, truly differentiates us. But another thing that really differentiates us is that we aren't a chat platform, as many of the tools that are out there are. We really looked at it very differently and we said, well, why?

Why are attorneys chatting? What are they trying to accomplish? And generally they're trying to accomplish some sort of work product they want to they have to then take whatever is in the chat and turn that into a memo they feel comfortable with, or, turn it into, motion. they feel comfortable with, or, turn it into, motion. What we wanted to say was, let's not assign the attorneys homework. Let's create the draft work product for them. So we are one click solutions.

Any drafting, any generative capabilities. You click the basically the the task, the out outcome, what you are trying to create. And from there we create it for you. You don't have to be a product engineer. We've worked in credibly hard, thousands and thousands and thousands of hours of our attorney analysts, reviewing with accuracy and tweaking and working with our engineers. And so, we wanted to take that work, the work of trying to make sure you have the absolute best prompt.

Leave that to us, trying to make sure that the results are accurate. Leave that to us, trying to make sure that the results are accurate. Leave that to us. We've done a lot of testing. You click a button and create the draft work product that you want to work with, and you're off to the races from there. Yeah. So instead of having to, formulate a prompt and then continue to, come up with follow up prompts to try and get it to where you want it to be, you click a button and it does what you want.

That is correct. And with many of those follow up prompts that you would need to do to try to get to where you are, we've done a lot of that work on the back end, so we have multiple chains going on. There's a lot of work that goes into what's the final outcome that the attorneys see. And so we, we want to make it where it's, it's, less up to how great a prompter, you might be. And more up to how do we, as a company deliver the best results in the most simple way to our users? Yep.

What's the best way for users to find out more about Trellis? I absolutely. You can go on to Trellis dot law and, request info, book a demo. You can go on to Trellis dot law and, request info, book a demo. We are here to help. You can also reach out directly. To me I'm on LinkedIn. Nicole. Underscore a underscore. Clarke. Or you can reach out to sales at Trellis dot law Well, congratulations on this launch. As I as I mentioned, I've seen a demo of it. I've written about it on my blog.

And I think it's really, really cool what you've got here. So, congratulations and thanks for coming on the show today to talk about it. Absolutely. Thanks for having me, Bob.

Transcript source: Provided by creator in RSS feed: download file
For the best experience, listen in Metacast app for iOS or Android