Winning Cases with Wexler AI: Gregory Mostyn on Why The Future of Litigation Starts Here - S10E25

0:00You're always marketing yourself in one way or another. I think litigators are going to want to prepare like an athlete preparing for the big race. AI that can help you play out different scenarios, using the facts, training up, putting someone on the stand, having a sort of sparring partner with a judge or arbitrator. Those are all the things that we're really interested in. The workflow in Wexler would be to build a chronology and then we even have a verify button where you go through and you verify each of the facts that are in that chronology at the point of finalizing it. That process of verification doesn't just make sure it's accurate, but it also makes sure you've got your fingers on the documents.

0:32So you're actually reading the documents themselves as they should be read exactly as they would look if you printed them out, which is really important. It might take three hours to do a chronology in Wexler. It doesn't take three minutes and it doesn't take three days. But those three hours not only help you get the best output, but they also help you become an expert in the case. On today's Legally Speaking podcast, I'm delighted to be joined by Gregory Mostyn. Gregory is the co-founder and CEO of Wexler. Wexler offers advanced AI-driven solutions for e-discovery, document review, and litigation services.

1:02With clients including the likes of Clifford Chance, Herbert Smith, Freehills Kramer, and Burge Salmon, Wexler's on a mission to solve the most complex cases. So a very big warm welcome to the show, Gregory. Thanks, Rob. Nice to see you. It's an absolute pleasure to have you on the show and looking forward to diving into all the great stuff you're doing in and around the world of law before we do that got a couple of serious questions what is your favorite beverage and what is your preferred choice of footwear on a typical workday beverage um i'm a sucker for diet coke i'm afraid a bit of a bad habit um dc as i used to call them um yeah i think i powered my university with a with a a sort of diet of diet coke if you like revising for finals where I got addicted so I try I try and limit my diet coke intake footwear very much trainers um I'm not sort of running around in Birkenstocks like some of the tech founders I know but uh trainers um but not not sort of suited and booted every day um unlike one of my colleagues who's a former lawyer who insists on remaining smart even in the kind of startup environment so yeah there we go like it bit of chalk and cheese and yeah I must agree with that, so I do like the odd DC as well.

2:14All right, before we get into too much of a rabbit hole, let's talk a bit more about you. Would you mind telling our listeners a bit about your background and career journey? Yeah, so I've had a peripatetic career, if you like. I studied at Oxford a while ago now, and I think the first thing you should know about me is that I'm from a family of lawyers. So my dad was a barrister and then a judge. My brother's a partner at Cleary. My stepmom's a barrister. I've grown up surrounded by the law. barristers you know every kind of dinner party my parents hosted was talking about this or that case you know Sunday lunches big debates about the latest and greatest in what was going on in the kind of world of law and how that you know more broadly impacts the world in general so that's the kind of personal connection to this in terms of me myself so I was very set on working in the music industry when I was at university I founded a few different events businesses and I thought I was going to become music mogul and went into that realized well it was actually great I mean I I learned so much and I very quickly rose up in a sort of events company, but realized there's definitely a bit of a ceiling there and wanted to sort of challenge myself a bit more.

3:19So I moved on to something which was sort of bridge between the tech world and the music world. And my thing was marketing. So I did marketing and sales, but all about kind of selling in a B2B and a B2C way. So my thing, I suppose, I'm good at commercializing, solving problems and thinking through, you know, what makes people tick, both B2B buyers and consumers. And so that's what I was doing in kind of the music world. But then I realized I had this kind of untapped entrepreneurial urge, which was from those days of running those events businesses at university, which kind of funded my whole way through without doing catering or tutoring or anything like that.

3:55And so I joined Entrepreneur First, which is an incubator set up by a couple of ex-Cambridge grads whose peers were all becoming consultants and, dare I say, lawyers and bankers. And they wanted people to be more entrepreneurial and think bigger. And especially in a UK context, like bring some of that, you know, we're a largely services industry and bring some of that innovation to the UK economy. And so I did that and realized that I had a sort of unfair advantage in the law and legal AI was kind of blown. Well, it was initially blowing up. It was basically just Harvey at that time. But I wanted to do something that was more specific, more specialized, which was really about facts.

4:36And this came from a conversation with my dad, which was about having the facts at your fingertips as a litigator is more important than anything. And as a sitting judge, he was, you know, seeing people standing in front of him, barristers, and they didn't know the facts in their cases, really. They didn't know the chronology. And so we set out doing that. And, you know, maybe my faith was tested at first. Maybe we should be broader. Maybe we should be doing everything. But actually, I'm very, very glad that we went deep. We went specialised and that's now allowed us to create something really durable and really deep that really can can grapple with these very messy litigation data sets and can produce work product that's kind of client ready, if you like.

5:11So that is the overview and a kind of short backstory to Wexler and me. I'm just sat here just nodding. There's so many parallels with myself almost hearing what you say. I came from a legal family that our listeners will know, my late grandfather, but it was my inspiration for getting into law. I at university was also running lots of events, club nights, and just had visions of me sort of handing out the flyers or promoting certain events, monster trucks and things like that, which I've been able to bring in recently into the law. And yeah, and then you obviously noticed you had that entrepreneurial flair and you've been hugely successful.

5:43So I love that story. Thanks for sharing it. And I love that, you know, I thought very openly about specific is terrific and being inch wide, mile deep. And you absolutely lazied that approach from your own entrepreneurial perspective. So as then the co-founder and CEO of Wexler, talk us through a little bit more what it is, how it works. The fact intelligence platform is Wexler. So that's a kind of phrase that I coined before. It's kind of like the exercise of establishing the facts in any dispute or investigation. So that's extracting the facts, i.e. trawling through millions of documents, pulling out the discrete observable happenings, if you like.

6:21They may be verified facts. They may be true or not. That's kind of for you to do with the platform. But basically the events mentioned, like Wexler's being founded in 2022 or the date of this podcast, etc. And then analyzing them against context. that's against the issues that might be from the claim form the complaint the pleadings etc and then verifying them so using the system to corroborate that with evidence and then turning that into work products so chronologies looking for inconsistencies and discrepancies using that to draft you know documents answer questions and so in practice where does this get used it gets used right at the genesis of a case so client self-produces 10 000 documents and says get back to me by friday with where we stand and there's four panel firms all in that process you know who can get to the facts quickest and best or it's for the sort of deep strategic layer the kind of focused review for trial prep we have a lot of barristers using the platform and plus witness you know the whole witness process deposition prep in the US summary judgment briefing and kind of anywhere in between and it's jurisdictional agnostic because as my dad says facts have no frontier and it's it can be used across arbitration litigation investigations so very simply we kind distill massive data sets no matter how messy they are the different file types the different formats the complex fact patterns we extract the discrete observable happenings from those events if you like we assign them relevance based on a list of issues that you supply and then you use that as the unit of analysis to create work product so yeah does that explain it very much so very much so and i love it you know when the facts are the facts pound the facts as they say for sure um i want to talk about the mission of wexler because i strongly believe that it's mission focused companies are the ones that are going to thrive and survive in this bubble so you know i think historically people have just sort of put values and missions and things on walls and they haven't really meant too much but i believe in modern entrepreneurship they carry more weight than ever and your mission i understand is on a mission to solve the most complex cases and you're going deep there but can you share the meaning why you went for that specific mission i think that it was funny i was talking to my uh again my dad about this um and in the prince harry versus the mirror group case they had to just take a very small selection of the of the kind of exhibits to to rely on at trial because of the overriding objective of like you know not wasting court time etc and it meant that actually they were only looking at a very slim segment of the total evidence which could be considered to resolve the dispute and i think that isn't necessarily just you know you're not looking you're not considering all the angles there could be a critical piece of information and if you're a client you're paying for something and you're not actually being represented to the fullest of the of the potential that you could be because there could be a critical piece of information that never sees the light of day because there's just too many documents to review and so I think it's about making sure that we're not limiting justice by just how many documents humans have the capacity to read and usually it's like paralegals reading them in the middle of the night or you know busy associates at 3am so i think it's about making sure that clients and you know in-house teams have the best representation and they can have their interests represented you know in in in the most in the sort of fullest sense not just about like the kind of reputation of the firm who's got more capacity to read more documents etc and then i think it's about um you know and this is actually a noble claim you know like I think ensuring those lawyers get to go home a bit earlier to see their wives and children and and not have to do the sort of drudgery and the grunt work um that that you know some people say it's kind of par for the course but actually we should be thinking bigger than that so I think it's about you know access to justice or maybe just you know ensuring that the full corpus of documents related to something can be analyzed and you're not just limited by the number of documents you have the capacity to read and also making the quality of life for those people and better two huge points and two things we absolutely support here on the leaf screen podcast is absolutely access to justice and tech for good i think the pandemic taught a lot of people the importance of you know we're only here once work-life balance genuinely matters and you know actually using this technology to have those you know hours i speak quite openly about um you know my late grandfather who ran a very successful law firm i remember this time quite vividly asking my mother um you know what your um memories of your your dad my late grandfather she said to be honest with you rob i don't have many who was always in the office yeah and of course you know that was that was the generation back then in 1950s running a law firm you know being competitive but actually if nothing changes nothing changes and with innovation you know change should be good and it should have impacts on society for good if we've got good actors doing good things and that's why we welcome great things like you're doing and other entrepreneurs we have on the show because i think you can make a not just within the legal profession but lives better and so that's why i think technology done right is incredible and i touched on before values i you know i see a lot of values we speak you know we've done 500 out of these episodes but yours really landed with me i love them because they're a lot of what i'm about so hopefully i'm getting these right hustle with purpose fail fast learn faster and one team yeah tell us about those so i think that uh someone asked me what's the one thing you look for in in um in new joiners and i do think it's this hustle with purpose piece so it's basically will you go out of the out of your lane to solve something even if it's not in your job description and will you hustle but will you do it in a way that's like you know disciplined and is not going to do anything immoral or illegal or you know that's obviously an extreme example but are you doing it with a kind of bigger picture in mind because we're all about that at Wexler when we don't want people who are like oh I've done you know I've been a sort of lead engineer for years and this is my way of doing things like our engineers when we were an early stage company used to come to conferences with me and do the sales pitching even though they had no they've never done sales before because I was the only person doing sales in the company and that's the kind of example of hustle with purpose kind of embodied so really hustling but making sure you're doing it in a deliberate way and not sort of randomly flailing around is super important fail fast this is something I learned as a marketer so it's a growth mindset you know it's better to to test something and it fail than to not test at all the silent killer is doing nothing right that's what one of my sales mentors told me back in the day and again a few caveats here we don't want to deliver something that's a poor product experience you know we don't want to put something else half baked into the world but I think it's a bias to action and to testing things rather than sort of strategizing and hypothesizing and one team I think you know we see a lot about we hear a lot about 996 and working and all that kind of thing and of course we work extremely hard we go out of hours we do things but like Kush and I are my co-founder Kush that is you know we're not straight out of Stanford we're not 23 years old like we have perspective we understand that actually our most valuable resources are human capital.

13:07And we want our team to be rested so that they can perform at the best of their abilities. And that means having, as you say, like a work-life balance and ensuring that you go home at the end of the day to your family and you get a bit of space and then you come back and hit it as hard as you can. And so the way we think about it is an elite sports team rather than a family. You know, we're not like family. We don't spend all of our time outside of work together. But when we come in, we're an elite sports team and we are very high performing and then we sort of have that um separation so so yeah that's a bit of uh elaboration on the values i think it's good isn't it that you have it that branded as that elite sports team because typically that is the bar and everyone's going to strive to deliver at that bar if they buy into the mission and they see that and so you know you keep that level um up there i think it's really smart and yeah i like the way that you um described that as well and and ultimately like you said in the current format and the world that we're living i think you have to be adaptable you know a job that could potentially turn itself on its head and i mean i run a legal recruiting business and you know traditional legal roles are going to be transformed and are being transformed so you know you're going to have to be a little bit adaptable particularly if you're an entrepreneurial tech focused business so i love that you actually have that really into the the core dna of what you're you're building and what you're about okay so you've talked to us about wexler but i want to get a chance for you to really sort of go deep on this because i think it's a fascinating um you know thing that you're building so talk us through in any granular extra detail on the sort of workings of it sort of those different stages because documents in fact extracted evidence connected then the contradictions exposed and a single factual view there's a lot to digest there so take it away yeah so i think it's probably the easiest way to explain this is is sort of with reference to an imaginary case so if you imagine like a multi-party fraud claim with um let's say there's 80 000 documents information scattered across board packs, committee minutes, emails between six individuals, late produced texts, third party reports.

15:07You know, as you'll know, things are privileged, things are in different languages. There's spreadsheets where there's, you know, data hidden in different sheets. It's like being overwritten by other documents, etc. And the claimant would say your client knew, the client says they relied on external advisors, advisors say they flagged concerns, the documents and this is the thing about litigation like they don't support any single narrative and it's very messy litigation data is very messy and so I think what Wexler does is we process this sort of process elucidates and rationalizes this mess because we extract the facts from those documents as discrete sourced units which are then deduplicated so you know we're not just looking at that mess of documents and asking the AI to sort of work out what's really important.

15:53And also it runs out of context. We've had issues, we've heard of other platforms like a long document, it just like stops on page 10. And so what happens is, we first extract all the relevant facts into those discrete units, and then use that as the unit of analysis. And so that's deduplicating. And actually, we've then got this massive index starting at the first year, let's say it's 1910 or something, and ending in 2026. And every single fact and a lot of them are deduplicated. And they're also given relevance against the list of issues. And so what this means is it doesn't just tell you sort of here's a general summary of the documents it tells you exactly what happened when it happened why it mattered and what it means for your case and and that's the kind of best way I can explain it really like litigation is very messy and you need something that doesn't just sort of run out of context sort of summarize it tells you in very concrete terms these are the facts that matter this is why I've extracted from the documents and that's why and I love that because that key point I always talk about just generally you know everyone's interested in their own favorite radio station what's in it for me and that last point about you know why it's important to them i think is so important so they you know get that absolutely love it okay we talked a lot about wexler as the brand and the organization let's talk about you you know as a co-founder and ceo what does the role look like day to day at wexler what are some of the main responsibilities and things you get up to so i was funny i was speaking to someone a friend who has just done an mba and he's like i want to do a strategy role at a startup and i was like there aren't strategy roles at startups either you're building or you're selling right and that is the thing and as a as the ceo my main job is selling the product and then obviously feeding back about all of the things that i'm hearing from customers and then taking that to the product team hiring is a key part of it but i actually have a bit of secret weapon there which i'll tell you about in a second and then obviously fundraising so those are like the core responsibilities right now we're in this massive growth phase like we're growing an insane amount.

17:45So my main responsibilities are selling, you know, to new clients, account expansion. So we have like a really high, we have over 300 % NRR because all of our customers want to expand with us all the time. So a lot of my, you know, day to day is working with customers about planning the expansions. And then yeah, networking and building up the book of business, plus thinking more strategically about bigger picture, what does fact intelligence mean in the wider market. So yeah, I think if you can't in a sort of direct sales environment like this, if you can't sell, you're not going to get very far.

18:19So, you know, selling is a huge part of my role. But then it's also selling to future employees and then selling to investors. Like you're always marketing yourself in one way or another. Today's episode is brought to you by Clio. If you're spending more time managing your practice than practicing law, it's time for a change. Clio is the intelligent legal work platform built for modern solicitors, combining context-aware AI with trusted legal research to help you work smarter, not harder. Switching is easier than you might think. Clio's dedicated migration teams handle your client and case data every step of the way, so you can focus on what matters most, your clients.

19:00Plus, award-winning support 24 5 via live chat phone and email means help is always on hand when you need it visit clio.com forward slash uk to find out why thousands of uk solicitors trust clio now back to the show mark cuban famous shark tank inventor says sales cures all and i think that's so true you know without sales you don't have any business however hard that hits people that is a fact of life it is the lifeblood, along with cash flow, which is always helpful as well. In terms of the journey, again, I'm really curious because you've obviously thought very long and hard about the mission, your values, the way you're going about growing, etc, etc.

19:40And when you co-founded Wexler, after seeing these firsthand lots of the inefficiencies of litigation, which you've referenced, you know, critical facts buried across all these huge documents, varying documents, etc etc different formats what are some of the challenges in the legal industry specifically you are seeking to address and you really want people to know that's where we need to go to Wexler for the challenges in the legal industry specifically there's a few things so one is there's so much more data being created now there's like email obviously was why e-discovery was created because suddenly there's so many more things written down there's so much more evidence to trawl through.

20:21Then we had instant messages. That was like the next sort of step change in terms of data volumes. And now we've got AI, which can generate anything, basically instantly, a lot of it is sort of slop that's irrelevant or is homogenized or whatever it is. So I think one of the main challenges we're solving is this kind of information overwhelm and trying to find the kind of winning thread when there's just so much data to trawl through. So I think that's a kind of key point. I think another thing is like, as the doc review tasks become less important, and let's say less valuable, because AI is doing them all, you know, it's not going to, let's play this forward five years time, or maybe that's too far, maybe two years time.

21:05You know, it may not be as much about who can find the killer smoking gun, because AI will help to do that. Obviously, there'll be ways of doing it faster and better, but it's about how you then use that and so the kind of human side of practice so mediations arbitrations um you know trials depositions i think lawyers litigators are going to want to prepare for those like an athlete preparing for the big race and so ai that can help you play out different scenarios using the facts training up you know putting someone on the stand having a sort of sparring partner with a judge or arbitrator those are all the things that we're really interested in um that's going to become even more important like the pressure is sort of counter cyclical it's going to go it's going to be even more important to do a good job on the day and the advocacy part is going to be even more important because it's going to be less about who can find the killer document or the killer fact it's going to be about how that gets delivered so i think that's a that's a key challenge like an emerging challenge which we're sort of very aware of yeah and clearly well at the heart of it as well and and thinking forwards for sure you touched on e-discovery it's where i wanted to go next really because you know you described earlier as Wessler as the fact intelligence platform which I really like by the way what does that actually mean in practice though how does that actually differ from the traditional legal AI or indeed those sort of e-discovery tools that are so many out there first of all we complement e-discovery and we you know we integrate with a number of the leading e-discovery providers and um you know what we don't do as much is is like this sort of first level of review, the cull going from millions of documents to 10 ,000 or 20 ,000, even 100 ,000 documents.

22:43What we do is we then turn that into actionable kind of case winning insights. So that means extracting the facts, like I've said, assigning them relevance, and then using that as the unit of analysis. So the thing that we do differently to eDiscovery is the data model of eDiscovery is rigidly built around documents as the unit of analysis. You know, they say, here's these documents are responsive to these search terms. What we do is we start with the fact, i.e. what's actually happened and why does it matter? And then we build up our scale based on that. So kind of a totally different data model and re-architecting it from an e-discovery perspective, you know, requires a complete overhaul and a reinvention of that data model.

23:19Whereas for us, it's about, we've got the quality and now we're kind of increasing the scale. But as I say, we integrate with e-discovery platforms and it's a slightly different use case, you know, it's not sort of the paralegal-led first level review. It's like the kind of barrister or the associate or the sort of trial attorney who's saying how do we then use that from a smaller universe of documents but but why does it matter and how is that going to help me drive forward my case and then um with generalist you know legal ai i think it's really about the um again it's about uh that kind of um distilling the documents into these facts no matter how complex and messy the different the data set is so it's it could be a like i'll give an example um generous legal ai it has a bias towards more structured information tends to be trained on corporate databases you know like it's very pro forma it's very kind of rote transactional work um you know the documents are sort of neatly laid out and it's kind of a lot clearer you know there's only so much variance and it's usually like a few sentences that have changed whereas litigation datasets could be like a whatsapp message medical record handwritten doctor's note an image a sort of scrap of paper a marginalia whatever it is and so um you need to do this pre-processing first to get the best insights and that's what you're looking for in the kind of complex theater of litigation like in very concrete terms the sort of vault and tabular review features they'll sort of summarize a document um as a single row in that platform and the document could be 10 000 pages long like it could be a whole trial bundle and the assistant will also run out of the sort of context to answer those questions so it's all about that that kind of pipeline we put the documents through to rationalize and elucidate the data sets into the facts and then use that as the unit of analysis and that is what helps you find the kind of critical piece of information even if it's buried on page 949 of 4062 right yeah and that is the reality right some of these things are huge.

25:22We talked earlier about challenges and you're looking forward. Let's just get into the mind of your general innovating approach here because there are lots of changes happening in and around the legal sector. So with the ever-changing technology, looking at the horizontals as well and what some of them might be doing, involving client expectations around billable hours, etc. How do you adapt as a business to market forces and changes to the legal landscape? It's interesting. I think we obviously always we have like customer advisory boards and we're always like listening to our customers and trying to understand how we can best position Wexler to work with them on this kind of journey.

26:02There's obviously rightly apprehension, but there's also a lot of excitement. So it's about making sure that people have the, you know, we host these dinners every quarter where we invite kind of both litigators and innovation folks and barristers. And we do them in New York as well. And we kind of have a roundtable about what this all means and how people are taking practical steps to stay on top of things. So, yeah, it's about being that kind of trusted advisor, training people. We actually do AI workshops for some of our barristers, which are not about Wexler. They're just about AI and something that people don't quite understand, which is like, how can AI be so good at solving a really complex reasoning problem and also not know the number of Rs in a strawberry, in the word strawberry?

26:42People try and explain how that works. So we position ourselves as this trusted advisor. We're always keen to discuss, you know, how people are thinking about billing and value based pricing and fixed fee arrangements and all those kind of things. Although obviously that's not, you know, it's not really up to us, but we will support how Wexler plays a part in that journey. But I think it's just about being like best in breed in terms of the quality of the output. So you're genuinely useful to the lawyers and then supporting them on the market dynamics and making sure they're across all of the big changes that are happening in the world of AI.

27:14So, yeah, it's about being that kind of trusted advisor. I love that. Yeah. And clearly you are. I want to go back also to when you were talking about your MBA friend now, because you're talking about different stages, you know, principally on the funding side of things. Because, you know, I think, you know, entrepreneurship is the biggest MBA of life you can get, getting out there, living it, being the experience, all of that good stuff, because nothing hits you between the eyes more than real life. You know, and you've experienced some great growth, you know, significant rounds of funding, you know, backed by, you know, Herbby C, Seed Camp, Legal Tech Fund, Mirriad Venture Partners, etc, etc.

27:44I think you've raised over$5.3 million in seed funding in 2025 alone. Congratulations. What have been some of the biggest lessons for you when scaling Wexler? That's a good question. So on the fundraising side, we learned a lot at EF. They taught us a lot about how to run that. I think it's a bit like life, really. You need to create competitive dynamics. like people unfortunately VCs are also you know like humans in that they like get FOMO and they want what they can't have and so you have to be really disciplined in running the process so that's been a big learning and something that I've I've learned a lot but I think also it's important not to kind of over optimize for a funding round like you need to just focus on building the best business possible and then and then good things will happen if you're growing fast enough in terms of scaling after you know once it was a bit of a it's a funny like mentality shift when you you go from being like basically profitable which we were before to then venture backed and that kind of mind shift like it's in my nature to be obviously be growing really fast but to like not spend sort of in a profligate way um but then you know that's a bit of a mind shift and a mindset shift which which we've definitely got our heads around now but I think every founder it goes through that thing when we like you've suddenly got loads of money in your bank account and now you're like right now we need to spend but actually our revenue is growing basically in line with our with our funding position as well so it's like we'll probably never spend even half of the money that we've raised that we raised um and then I think on team from a team perspective like we we're trying to keep as much as possible a kind of flat structure um we don't want to be bogged down with you know like multiple levels you know tons of internal meetings um I'm sure I think people come to an early stage startup because they have real agency and real ownership and they don't want to be sort of stuck in strategy sessions and meetings and those kind of things.

29:39They want to be either building or selling, like I said. So yeah, I think we're kind of at a point now where the next level up is going to have to require a reorganization. But right now we're operating at an extremely high level and kind of bringing the best out of everyone. So we're trying to keep it at that point before we sort of inevitably take the next step up, which is going to involve hiring in the us so we've got like a sub entity in the us and we're hiring in the us and getting boots on the ground at the moment so i think that's going to be like the next level very exciting and as i always say to entrepreneurs you know new level new devil right there'll be new exciting challenges opportunities that come as you keep going through the the levels of the journey um in a recent rexler article titled hallucination the dirtiest word in a you explore why hallucinations shouldn't stop lawyers actually using ai so would you mind telling us a bit more about hallucinations you touched on it previously and how they actually can be mitigated i think this article was inspired by a post by ben evans who's a brilliant uh technologist who has a really good perspective on ai like i think it's a healthy amount of skepticism combined with how powerful this technology is i think his his his article was basically that anyone who's completely cracked the accuracy problem with ai is lying to you but that doesn't mean it's not incredibly useful so yeah i think you need to have both of those things in mind when you're using ai so wexler doesn't have like fabrications so what i would say is a fabric hallucination we like to say internally is like if a human given the same information would not make this mistake like a kind of intelligent human then i would call that hallucination like it's an unexplainable situation where there's no sort of source etc and so that Wexler doesn't have that issue.

31:20Is it 100 % accurate? Of course not. And any AI founder telling you it is, is, you know, you need to sell them back the bridge they've just sold you. It's really a question of like being able to very quickly independently verify the output, checking it, using it, and then using that as you build up your output. In terms of practically how we managed to mitigate that process, I think, I know I keep coming back to it, but it does come back to this like fact extraction pipeline. So if you're not asking the models to do too much at one go, then you can massively mitigate, you know, context window issues leading to things kind of being invented or fabricated.

31:55So we sort of say A to B to B to C to B to C, sorry, to C to D to E to F, etc. And stitch together rather than A to Z in one go, because in doing that process, you're building up this kind of curated, accurate database, where it's not sort of having to grasp in the sort of fumble in the dark to find a piece of information which may not exist. Plus, we have a huge amount of guardrails on the back end, which is like only stick to that, you know, as you would expect. And that's kind of what you can do without that kind of custom architecture. Plus the fact that everything in Wexler is sourced down to the sentence level and is, you know, easily verifiable within the platform.

32:31So the workflow in Wexler would be to build a chronology. And then we even have a verify button where you go through and you verify each of the facts that are in that chronology at the point of sort of finalizing it. and that process of verification doesn't just make sure it's accurate but it also makes sure you've got your fingers on the documents right it makes make sure you're actually reading the documents themselves as as they should be read you know it you know exactly as they would look if you printed them out which is really important so it might take three hours to do a chronology in Wexler it doesn't take three minutes and it doesn't take three days but those three hours not only help you get the best output but they also help you become an expert in the case and And so that's why it's important to have that kind of process.

33:11It doesn't mean it's not unbelievably helpful and a huge amount of time saving, but it's actually important because you're becoming more up to speed and more familiar with the documents. And maybe even if you'd read them cover to cover, because reading them cover to cover, you get lost in the story of each document. You're not looking at things in terms of the fact patterns. I love that. Yeah, and trust the processes as well, right? And, you know, we talked around AI generally, you know, Horizontal's touched on a little bit. But, you know, it's very clear Wexler is specifically, right, built for disputes, not general use, legal AI usage.

33:48So why is specialized AI now our future? There's so much value from generalist legal AI and generalist AI in general. it's unbelievably useful you use Claude internally like the productivity gains you get are just unconscionable from what it would have been even a year ago you know it's it's complete level up but in the theater of litigation you're looking for the upper hand against the other side and you're looking for anything that will give you that advantage and and not just that you want something that understands the nuances of disputes you want you understand that things aren't always as they it's seen things are obfuscated or unclear or people are speaking in code words and you know investigations people are like saying very odd speech patterns those kind of things is where you need something that's bespoke and custom built for that challenge which is kind of really deeply understands litigation isn't retrofitted and has this way of distilling documents into facts and using that as the unit of analysis and in litigation if it gives you the upper hand, it's worth, you know, millions, because it could you could be in 100 million dollar lawsuit, and there's a huge amount at stake.

34:55And so that's where we're seeing a lot of pull from the market, people will generally already have a generalist legal AI platform, they may also have ChatGPT Enterprise or Claude or whatever it is. And they're using Wexler, because it understands litigation, it can do a huge amount, so we can do 250 ,000 documents of LLM analysis, which is, I think, more than any other platform I've heard of, like in terms of the deep LLM analysis. And it also will give you that upper hand in the case. Yeah, no, absolutely. I think you've justified that very well. You've mentioned US expansion. What else does it look like in terms of the future for Wexler?

35:34You know, five years is too long a time to say in modern worlds. But, you know, looking at the roadmap, you know, the coming years, where do you envisage wexler going next we think we've built in fact we know we've built something really unique um that is people see a significant delta against these platforms and so we just want to build up that advantage we want to build the best software for building the best case for litigators for investigators we're exploring some other verticals like claims handling insurance regulatory stuff we're actually seeing quite a lot of pull from enterprises for doing some of their in-house litigation.

36:07So yeah, it's just about consolidating, doubling down, increasing the scale to ingest millions of documents in a sort of in this same pipeline, and just building more of these workflows that can help bring out the human side in a dispute. So what I was saying earlier, you know, prepping for the big day in the same way an athlete preps for the big for the big race, and kind of searching for that, you know, identifying every critical fact that could help bring your case together. So yeah, doubling down, geographical expansion. Obviously, the US is the biggest market. We have a lot of usage in APAC as well, Australia and Singapore.

36:41There's a huge amount of arbitration and litigation there. And yeah, just sticking to our guns and being a scalpel, not a Swiss army knife and not losing our focus, basically. Yeah, absolutely. And that hustle with passion, right? I love it. So look, it's been a masterclass. Really enjoyed it. You know, lots of great lessons, learnings, you know, ambition right throughout the conversation. So thoroughly, thoroughly enjoyed the chat. Just before I let you go, finally, what would be your advice to aspiring lawyers who are interested in pursuing a career in law and indeed AI? I mean, I wasn't a lawyer is the first thing I'd say, but, you know, we've got our head of commercial strategy who is a litigator at Mayor Brown and he became the sort of firm expert on AI before making the move.

37:18I do think, I mean, I've heard a few people say this, but I think it's valid. Like if you are entering the sort of world of legal practice and you already are an expert in AI, like we've got an operations associate who's starting a training contract next year. And he is an absolute whiz at Claude and all the other models now. He is going to command a huge amount of respect and gravitas in the firm, even with much older partners. And so you can enter the world of work as an expert. You can enter the world of work with authority that you didn't have before, because before you were, you know, the lowest on the food chain and now you're the highest.

37:51I heard another anecdote of someone at a firm who built a sort of self-coded or vibe-coded application, which has now been rolled out across the whole firm when he was a trainee. So I think become an expert. You don't have to be like technical to do this anymore. Just become, stay incredibly inquisitive and curious. Make sure you have a healthy dose of skepticism always, but make sure you are the authority. So when you walk in on your first day, you can speak from a position of real knowledge and deep, you know, command some actual respect with people who are 20, 30 years your senior, because you understand this, and they don't.

38:26And this is what's happening in the world. Right? That's what I would say. I couldn't agree more. I love that advice. I talk very openly about the reverse pyramid in terms of the talent perspective, actually, historically, it's been like that, obviously, the more senior you are. But now, like you say, coming through, if you put that time and energy, and you understand these technologies, you're growing up with them, you're going to be a hugely valuable asset to any organization. So love it. Yeah, absolutely brilliant um Gregory if our listeners I'm sure they will want to learn more about you or indeed Wexler where can they go to find out more feel free to share any websites any social media handles we'll also share them as episode for you too very easy it's there wexler.ai you can request a demo you can email me directly you can contact me on LinkedIn no problem Gregory Mostyn I'll definitely get back to you um but yeah we're always happy to have a chat about what's changing in the world of disputes and AI um and obviously you can set up a demo as well so yeah Wexler.ai, very easy to remember.

39:16Perfect. Well, thanks so much once again, Gregory, for joining me today. It's been an absolute pleasure having you on the show. From all of us on the League of Speaking Podcast, sponsored by Clio, wishing you lots of continued success with your entrepreneurial pursuits and more. But for now, over and out. Thank you for listening to this week's episode. If you like the content here, why not check out our world-leading content and collaboration of the Legally Speaking Club over on Discord. Go to our website, www.legallyspeakingpodcast.com. There's a link to join our community there. Over and out.