What's Up with Tech?

AI Assistants in Law: How Lex Machina's Protege is Transforming Legal Strategy

Evan Kirstel

Interested in being a guest? Email us at admin@evankirstel.com

The legal profession stands at a transformative moment as artificial intelligence reshapes how attorneys work, strategize, and win cases. In this illuminating conversation with Carla Rydholm General Manager at Lex Machina, we explore how legal analytics and AI are fundamentally changing litigation practice while preserving the irreplaceable human elements that define quality legal representation.

Carla shares the fascinating evolution of Lex Machina from a public interest project between a law school and computer science department to a pioneering legal analytics company now owned by LexisNexis. Their mission revolves around "data-empowered lawyering" – transforming the overwhelming volume of legal information into structured, accessible insights that attorneys can use to make better strategic decisions.

We dive deep into Lex Machina's AI assistant "Protege," which allows lawyers to ask natural language questions about litigation timelines, judge tendencies, and strategic options. Rather than replacing legal expertise, Protege amplifies it by creating space for attorneys to think deeply and strategically about their cases. As one knowledge management director revealed, having instant access to data like a judge's unusually high reversal rate can prompt nuanced client discussions that ultimately lead to winning strategies.

What emerges is a vision of legal practice where AI tools become as standard as word processors, helping attorneys manage information overload while focusing on the judgment, client relationships, and ethical considerations that cannot be automated. Despite the technological evolution, the future of law remains fundamentally human – enhanced rather than replaced by artificial intelligence.

Ready to explore how AI is transforming one of our oldest professions? Listen now to understand the sea change happening in legal technology and what it means for the future of justice, representation, and advocacy in our increasingly complex world. Then share your thoughts on how AI might reshape other tradition-bound professions!

Support the show

More at https://linktr.ee/EvanKirstel

Speaker 1:

Hey everybody, Excited to dive into the world of legal tech and how AI is being used in the legal profession today with a true expert innovator at Lex Machina. Before that, maybe introductions Carla, how are you?

Speaker 2:

I'm very well. Thanks for having me. How are you? Are you?

Speaker 1:

I'm very well. Thanks for having me. How are you? I'm well, thanks. Thanks for joining. Really intrigued by the work you're doing. As someone who's definitely an outsider with no legal background at all, I would love to learn more about you, your mission and what's the vision at Lex Machina.

Speaker 2:

Sure. So Lex Machina's focus is legal analytics. It's what we do, it's really our mission. It is to bring legal analytics, this idea of data-empowered lawyering, to all areas of the law. So we got our start as a public interest project.

Speaker 2:

It was this idea between a law school and computer science department that was funded by companies and law firms and then spun out as a private venture in a law school and computer science department that was funded by companies and law firms and then spun out as a private venture were acquired by LexisNexis 10 years ago and have been doing ever since like we're building the capabilities that were in some of those early pitch decks, sort of living the dream, even it's. It's uh, even uh, surprising to me, but I've been here for a long time. I'm now the general manager, but I got my start because I loved working with data scientists who could really use algorithms, use technology, mapping opportunities to apply technology to different problems, and so I'm an attorney by training and have sort of a stats and science background as well. So the idea of creating data sets it resonated for me and I'm still doing it and leading the team. It's a very integrated team of attorneys who now work at a tech company, and then data scientists and engineers.

Speaker 1:

That's fantastic. You've had quite a 15 year or so journey, I think, through this whole process involved in data science and analytics, before it was so exciting today with AI and Gen AI. So you know, fast forwarding to the present, where, would you say, ai is currently in the legal industry and where are we in terms of a sea change or turning point in the industry? And where are we?

Speaker 2:

in terms of a sea change or turning point in the industry. I love the sea change. That's a really good, a very vivid metaphor in that I think it's a really interesting point and that the industry real diverse efforts going on where attorneys and teams are seeing opportunities to use AI technology in their current workflow, define new workflows, using AI both as a tool but also as a brainstorming, you know, sort of like a digital colleague, so to speak. So there's lots of discussion in legal and tons of activity in terms of how AI is being incorporated and what's the plan, you know, at the firm level, within in-house and corporations, and then definitely the court systems are thinking a lot about it and really providing a lot of opportunity to discuss like, where are we headed and what's the plan?

Speaker 1:

So interesting questions there. So my limited interaction with lawyers, the legal team, seems to be pretty risk adverse and kind of wedded to old school technology, in my little humble opinion, in terms of loving yellow legal pads and fax machines and like old school apps and things. So you know how are attorneys and firms your customers, partners reacting to your new suite of AI tools?

Speaker 2:

Well, I definitely appreciate the pairing of just thinking about legal and industry and just sort of you know, definitely tradition and definitely being cautious, I think attorneys are expected to really weigh the risks and be very methodical.

Speaker 2:

The industry is also known for being very analytical known for being very analytical. So I think there's been a shift as the opportunity is really recognized, whether it's coming from partners that can see that everyone's being asked to do so much all the time at our work. So, if there's opportunities to streamline, help them keep thinking deeply about insights for their teams, if there's really that opportunity. You know people don't want to miss out on something that's good for them. So whether it's good for the teams to stay focused on the people part, you know law is also a very team based profession and really the sides of humans that will keep always doing. You know, practicing judgment, connection building, like that.

Speaker 2:

Ai isn't replacing that. It's more this really neat moment where technology applies with a real resonance for legal, and so there is this appeal to using AI assistance to reduce friction, carry the workload. So it's just the right place, right time. Legal is a great application for AI because there is just so much information, so many documents you mentioned legal pads just the volumes of information that teams are trying to reconcile and really that sort of a take control attitude that maybe one also associates with lawyers are seeing that it's sort of on their terms and just getting involved and deciding how they're going to use AI. So I think definitely the caution and precedent holds true and it's a very, very methodical approach and there's a lot of discussion on ethics, a lot of discussion around the implications for how our team's going to adjust, because things are shifting quite rapidly.

Speaker 1:

They are indeed, so let's talk about your AI tools. Protege, I believe, is a brand Great name. What is it? How does it work and what was inspiring inspired to build it? What was the big idea behind the tool?

Speaker 2:

Oh well, as I've mentioned, lex Machina's users are attorneys. So at law firms, at corporations, and really our users are asked to consider many kinds of questions all the time and make decisions throughout the litigation process. For example, who should my company hire to represent us in a new case? It's what an in-house attorney might be thinking about. How long will this litigation process take? You know that's like. What are we in for? You know what's the plan. Should we fight an unfavorable decision? So there's questions and really seeking to make decisions over and over um, so protege and like smock and uh, provides customers with insights, so actionable insights that help them answer questions, and we're doing this because we've created these legal analytics data sets.

Speaker 2:

To talk about my tenure here, you know but, but data and court documents, really transforming information that's out there. This is like complex litigation records into very structured data and once you've got that, really the now the experience of accessing the analytics and insights, it's easier with the sponsor protege and that you you know our users ask the question just in, in natural language, and those insights. So how long? How long will the litigation process take? Type in the question and then this assistant takes you to the analytics where you can see the information. Think deeply about. Well, in this jurisdiction, the median time to trial is this number of days. I need to make sure that my client is aware you know that it's going to take this long. So it's really helping helping meet that moment that there's so many questions to get answers to and just make it make it easy. Make it easy to access the data. It's, it's structured now in our database and really it's. It's that sort of simple as like help our users answer all the questions they're trying to get to.

Speaker 1:

That's wonderful. What a great approach. So easy is definitely a key design principle, Jack. Gpt is easy, but of course, how do you then build in trust into that product so they can rely on it? You know day to day in a real world environment.

Speaker 2:

Oh well, and that's where it's protege provides the answer and that's where if I expand on that just a little more that it's really this jumpstart into getting into the data and, once the answer is taken, the ability to get in deeper to really understand, and there's more going on there. But really we do provide an answer and then sort of that opportunity or forum for the user to like well, you know, I really want to understand this answer and kind of get you know, sort of have their own independent thoughts and conclusions around the answer.

Speaker 1:

Yeah, I get it Totally. So you know helping with productivity is important, it's one thing, but you know helping with productivity is important, it's one thing, but you know everyone's outcome driven. So how does it actually, or can it actually, help lawyers win when you know when more cases you must have some interesting anecdote stories on making a difference.

Speaker 2:

Well, I think, yeah, and a big part of having more, yeah, essentially having the AI assistants, ai colleagues, a big part of it is, yeah, this ease of access and really helping people have more time to be thinking and thinking about those insights to then have the best answer. So, in terms of thinking about how exactly is protege helping lawyers win, you know, sort of having an impact on decision making and strategy. It was a really interesting experience at a conference, at one of the legal week conference presentations where I helped moderate, and it was a really interesting panel a mix of law firm partners.

Speaker 2:

director of knowledge management and just the cred that the knowledge management director, who's coordinating all the technology and workflows and really having that vision, was really interesting because that's someone who's at the hub, you know, really helping the teams work efficiently, get results for their customers and some of the stories she told really around winning had to do with really providing that opportunity for the insights and the lawyers being given like more space and just like it's surprising, you know, we're all so busy answering email, dealing with one another that having that time and that space to really laser in and see opportunities and be tactical it went a long way and so some of it. It sounds like save time, save money, but it's.

Speaker 2:

and then having time to think deeply, and so some of the examples of winning had to do with how to have a more nuanced client conversation or really see the bigger picture by making use of all this data that's out there.

Speaker 2:

And so that's where having insights into, well, this judge's reversal rate on appeal so this is where if you lose or have an unfavorable decision, you can appeal, and just hearing that it was a very profound thing to very easily share.

Speaker 2:

Well, this judge's reversal rate is really high, you know, much higher than other judges in this court. And just kind of that shift of oh, maybe we should, it's expensive to appeal, but maybe we should talk about it and kind of just that like almost like reducing, like the unknown and that friction to then get to, well, what are we going to do, you know, and then setting course on a winning strategy. So I know that was a long-winded answer, so thank you for you, but it's it's really I'm trying to bring to life like what's going on on teams, and some of it is just sort of that relief you feel when you have something to react to and you're like, oh well, that's a really, you know, that's a really interesting piece of information. That it's. It's incredibly helpful to charting the best path and getting the best outcome for my client.

Speaker 1:

That's great, no great to hear and you see AI tools like Protégé becoming almost standard in every law firm in the next few years. The way Microsoft Word or Docs are pretty standard. It's hard to imagine, in a few years, working with a lawyer who doesn't use a tool like Protege.

Speaker 2:

Yeah, I do think that teams, that legal teams, whether it's the law firm context or in-house the corporate legal departments they will have really incorporated the AI assistance sort of their AI colleagues as part of what they do, because, with AI really being in a position to connect multiple processes and in some ways, anticipate when the team needs data via something like protege or really building an assessment for an early case. If there's a way, for example, that your firm, you know, say you're um a top lawyer, um evan, and you're, you have the opportunity. It's a brand new case. You just want to understand the lay of the land. There's a lot of inputs that right now, um, you would want to know about.

Speaker 2:

Well, this judge, opposing party, what happened last week, is there anything else? And if that was just really readily available to you without having to send off, you know, open a bunch of different things and have your team print stuff out, if it was really managed in a really coherent way and delivered in this, like anticipating the agentic flow, you could see how you, you, you can apply your own creativity, just like a lot sooner, instead of this like waiting to get this and that and kind of that, stop, start, and so that's where I think, with the standard presence in law firms, it's just natural to have something to react, to, have, um, really, uh, have that first drafts of a memo, have a digest. Someone drops off a box of documents and you're like oh well, I want to get that through our system as quickly as possible, and I think it's not a reliance on the technology, but it's like using it on your terms is what we'll see?

Speaker 1:

That's fantastic, and I think there's probably a whole generation of younger, up-and-coming lawyers who embrace this perhaps more than their older, more experienced colleagues as well, so I imagine a generational shift. But where do you see Pranush headed over the next year or two in terms of features capabilities? What can you share with us? I'm sure there's a lot you can't share, but whatever peek into the future would be interesting.

Speaker 2:

Yeah, no, I appreciate the question. I think a big part of some of the themes you've touched around responsible use of AI and whether a profession really embraces something that is a shift will have to do with or see change. I love that is really that we'll work really closer with customers and their journeys, with using the technology and the feature and really we're seeing really great usage in terms of engagement and it's sort of just like a natural fit, but really wanting to have an understanding of how the data and analytics are fitting our customers and users' needs and doing that sort of like the next rev on product development and really we want the product to anticipate how the user wants to interact with questions and get the data they need and really on their terms. And and we'll be doing a lot of looking at data and thinking about how to how to improve performance always, but also talking with customers like a lot and and getting users sort of their stories, um, but also their vision for how they want tools to function and how they want the experience of ideation and and really that insights.

Speaker 2:

You know we're I'm excited about bringing data to the practice of law, not because I think the data has the answer. It's like the attorney has the answers. But that too, when they see a number or they see a trend, will have that perspective as an expert in that area, as an expert in that area. So that's where really knowing where our users, where our customers, are headed in terms of their vision is a big, big, big part of it. So that's it's a kind of vague answer in that, but really what's next for Protege is we want to anticipate what users want and need, but also ask them like what do you want and need?

Speaker 1:

That's amazing. It's very analogous to what's happening in healthcare. My doctor recently mentioned he's going to start recording, like ambient recording of conversations doctor visits of conversations, doctor visits that will be fed into a private LLM and used for, you know, the EMR. So it's just amazing how all these industries are changing. Let's move into a little bit of the fun science fiction side of the discussion. If we can do a vision in five years, you know, maybe not in the US, maybe Dubai or Shenzhen or somewhere Digital lawyers, digital judges, ai lawyers or is this outside of your realm of comprehension at the moment?

Speaker 2:

Well, no, I've talked to judges about how they want to use tools and where they they feel the opportunity, but also have questions when they hear about full, full AI attorneys or law firms. There's the the first AI law firm in the UK was announced. They're very trying new things. The AI first was announced. They're very trying new things AI first. But I do think that the practice of law is really about providing access to justice, representation for people, for companies, and a big part of the practice of law I mean it's a uniquely human industry and so just that, the judgment and sort of connecting and really advising clients. I mean, I think the practice of law will stay focused on that client focus and there's also just a lot of focus on ethics, so I don't think that'll be automated.

Speaker 1:

Well, that's good Good news for those law students out there who might have concern. Of course you got to pass the bar first. I wonder if AI could do that. Maybe they could today, but fascinating world. Thanks for the sneak peek behind the curtain and keep up the amazing work so interesting.

Speaker 2:

Well, thank you for having me. It's, it's fun Thinking about where we're headed and just, yeah, taking time with you to pause and reflect a little. So, yeah, I appreciate you hosting.

Speaker 1:

And thanks everyone for listening, watching and sharing and check out our new TV show as well, now on Fox, business and Bloomberg. See you there, take care.