Pioneering legal AI startup Eudia has bought another ALSP, this time US-based Out-House. Artificial Lawyer spoke to CEO, Omar Haroun (see in-depth interview below), about Eudia’s AI strategy and what the iconoclast is trying to achieve. Hint: AI will, in time, totally change the legal world and they want to be right in the middle of this transformation.
One key takeaway is they have experienced huge growth this year as they target Fortune 500 companies’ contracting needs. A second point is that Haroun truly believes that the legal sector is going to be transformed by AI – and wants Eudia to be instrumental in helping that change to happen. That said, right now he doesn’t see many lawyers or firms truly embracing this evolution in earnest – which is one reason for Eudia, i.e. they’re going down the transformation road at full speed, even if some more traditional firms are scared to do the same.
And a third key area is the core belief – long shared by this site – that when it comes to ‘routine but complex’ legal work, which is highly amenable to process improvement and the use of AI, then the billable hour just doesn’t fit and that there are better ways to charge for the value a legal business such as Eudia delivers. (More on all of this in the AL in-depth interview with Haroun.)
Now onto the latest deal, which follows Eudia buying ALSP Johnson Hana, gaining $105m in a series A round, and setting up a law firm in Arizona. Out-House is a legal services provider founded by Lynden Renwick, an experienced lawyer and former GC. It will add to Eudia’s overall capabilities, which as Haroun explained to AL, is all about building a ‘legal brain’ for clients, which meshes together AI and human skills, in part to handle areas such as contract review.
Out-House exclusively services Fortune 500 legal departments, where it acts as an ’embedded extension of the in-house legal function’, they said, and focuses on ‘scalable support across high-volume commercial contracting and outside counsel management’. So, a neat fit.
They also noted that Eudia and Out-House have been working together for several months already, so the move makes even more sense. Moreover, as Haroun told this site, Renwick shares the same world view, namely that the entire field of knowledge work is going to radically change because of AI – and they want to be part of that change.
Renwick commented: ‘AI transformed my practice as a lawyer. Over the past two years, I’ve tested just about every tool in the space—most of which were clearly designed for law firms, then retrofitted for in-house teams where they just don’t work. Eudia stood out as the only AI-native platform built from the ground up for in-house legal teams.’
And here is the in-depth AL interview with Omar Haroun. AL would add that if you really want to understand what’s happing now in the wider field of ALSP transformation because of AI and the growth of ‘AI hybrid law firms’, then have a listen as we explore Eudia’s wider strategy and its ethos – as well as hear some additional detail about how they charge clients and how they operate.
Press Play to watch / listen inside the page.
We explore:
- AI is not just about software; it’s about transforming labour
- The legal industry is operating under outdated business models.
- Eudia aims to build a ‘brain’ for each company they work with.
- The integration of AI can lead to significant ROI for clients.
- Human expertise remains crucial in the legal field despite AI advancements.
- The current model of outsourcing legal work is flawed.
- Eudia’s growth has been exponential, with a thousand percent increase this year.
- The focus is on creating a new ecosystem for legal services.
- Knowledge management is key to leveraging AI effectively
And here is the full AI transcript. As I think you may agree, Omar’s view of the world is one founded on the idea that AI will be truly transformational to the field of knowledge work, and that includes lawyers. AL would 100% agree that is the case. Read on.
—
Hey, everybody. Richard Tromans here at Artificial Law doing a special interview with Omar from Eudia They have just bought OutHouse.
Very, very interesting deal and perhaps even more interesting is the bigger strategic picture of what Eudia is doing. First of all, let’s talk about the acquisition and thank you for being on the show. Just tell us a bit about why you did the deal and what’s the two plus two equals five aspect here.
Omar Haroun (00:47.199)
Yeah, so I’ve come to believe that most people think that AI is the future of software. We believe that AI is the future of labor. And when you look at how most corporate legal departments function today, 95 or up to 98 % of their budget goes to humans, services, people internal and external. And so our view from day one was that no amount of putting AI on top of the 2 % is going to really fundamentally change things. You actually do need to have an answer to what do the labor systems of the future look like in a world where most of this work can now be automated. And I think, know, personally, I’m actually very optimistic that I think there’s going to be more jobs than ever in this new world. There may be some temporary changes and transitions that we have to go through. But what’s really guided not only this acquisition, but a lot of our strategy more broadly is when we find the rare, I’ll just call it like less than 1 % of service providers or lawyers who are out there that are actually already living in the future and embracing AI, completely transforming how they do the work, then it becomes extremely attractive to partner as deeply as possible with those people. so Lyndon, the founder and CEO of OutHouse,
is an example of one of those people where we met him actually because he reached out and was kind of obsessed with adopting AI to transform his practice. And so he had been an MLA, a 100 lawyer. He then became a general counsel of a prominent company. His own story is quite interesting. The reason he left that company was he found because he was adopting technology, he was able to get through things like contract review five times faster than anybody else. And so he set up his own business and suddenly that took off where he was speaking to other general counsels and because of his adoption of technology gaining a major advantage. And so it just felt like a very, very good alignment between sort of how the two of us see the world. And at the end of the day, I don’t think we can change this industry without having the world’s best lawyers as part of the equation. So that’s about the rationale.
Richard Tromans (02:59.954)
Gotcha. And just to clarify, technically it’s a law firm in Maryland, but it’s kind of an ALSP as well, effectively. What actually is it in real terms?
Omar Haroun (03:12.561)
Yeah, so they had been a law firm, but they transitioned. So the legal entity is an ALSP.
Richard Tromans (03:21.948)
So you can buy it, so it’s no big deal for you to buy that. I was just thinking, because I know in DC there are special rules where you can actually do something like this, but I thought, Maryland, that’s slightly different. So, gotcha. Okay, well, that’s great. And I guess that builds on what you did with Johnson Hanna, which is definitely very much quite large, ALSP. And then of course you built your own law firm, which is technically a real law firm out in Arizona because of the bar rules there.
Omar Haroun (03:47.316)
Mm-hmm.
Richard Tromans (03:50.098)
which will help you to do some of the reserved activities as we call them in England or regulated activities. Let’s not go into every single item there, but just let’s just talk broadly as you know, got you for a few minutes on the show. I mean, what is the strategy here? What are you trying to see? Where are you trying to get to?
Omar Haroun (04:09.799)
Yeah. So the overarching strategy is that right now we have $6 trillion of human labor, which is broadly speaking, the professional services industry, which is still fundamentally operating under what we believe is now an outdated view that judgment can be packaged into time and essentially either billed by the hour or provided under some time and materials type of business model, right? And that doesn’t serve our customer very well because now that knowledge and judgment has gone from something that was a scarce resource to arguably, if not infinitely available, certainly much more accessible than before with AI. We think we have a great business opportunity and really opportunity to add a lot of value to our customers who have always been a hundred percent in-house teams by really just offering a much better alternative to that, which doesn’t offer that flawed business model. So I think I’ve, I’ve talked a lot about the billable hour in the past. don’t, don’t, I think people get overindexed. Yeah. But, like, like just, just one, one of the things that I’ll point out is like, I don’t think it’s just sort of the, the extremely high ROI that you get as a customer by going with a incentive model that’s aligned where now the provider is.
Richard Tromans (05:15.016)
…. the two of us could talk the legs off a chair [about billable hours] honestly, don’t worry about it.
Omar Haroun (05:35.977)
fully incentivized to use AI to get all the benefits. I think the other part that’s even more exciting to me is we’re really building a brain for each company that we work with. And the company owns the brain. We don’t own the brain, but our platform is used to develop it. And really that brain is a company brain, which in some sense takes all this knowledge. Cause I think when you outsource your work to a law firm historically, you’re not just outsourcing your dollars. You’re also outsourcing your knowledge. And the more you use that law firm, the more aligned you become on them.
And same thing with service providers or consultants.
Richard Tromans (06:06.728)
and you could argue kind of dependent but not necessarily in a nice way.
Omar Haroun (06:13.855)
Yeah. I I think there’s amazing lawyers at law firms and we’re working with any of them and all of that. So I don’t think it’s anybody’s being malicious, but I think the entire model sort of does, like when you think about it, the human expert externally is training their brain on your company’s data and then renting that back to you at rates that go up 18 % year over year. Right. So it’s not a great situation in the status quo where
Richard Tromans (06:17.992)
Yeah, sure.
Omar Haroun (06:41.543)
At least the one who’s really not winning from this setup right now, I would argue is the in-house teams. And so we believe by shifting the model and actually encouraging and promoting the expert to share their knowledge as part of this company brain, everybody can win, but it’s going to require a very, different kind of ecosystem and business model.
Richard Tromans (07:00.52)
Well, that brings me to the next question, which is unlike many legal tech companies, there are a few, there are a few, but you actually have a real body of sort of paralegals and actual expert lawyers altogether who are going to be doing, you know, that human judgment aspect in sort of unity with the, with the tech. Is that, is that a, I guess, because of necessity, at least that’s what I see in other legal tech companies, is a degree of necessity to get to that 99.9 % kind of, you know, quality level that the client wants. And then the second point is there’s a degree of owning the means of production, whether that’s intentional or simply because legal teams say, look, if we’re going to outsource this to you, we’d rather you just did the whole thing from top to bottom, sent it, you know, sent it back all wrapped up with a little bow on it. And if you’re going to do that, you need people, you can’t just have a tech solution. I mean, you just talk us through the logic of having tech and people on your side.
Omar Haroun (07:59.261)
Yeah, definitely. So I think most other AI companies in Silicon Valley have some version of AGI as their North Star, right? And that’s never been kind of universally defined, but Sam Altman once defined it as machines replacing humans at most economically relevant tasks. And we not only think that that’s a depressing future, but also not even a realistic one when it comes to domains like legal, because ultimately trust is a human to human sort of experience and ultimately the accuracy does need to be 100 % or at least people believe there needs to be 100%. And so we’ve just made our problem statement very different from AGI, which is can we prove that one lawyer can do the work of a hundred historic lawyers and actually start to do a lot of work that was never possible before because they now have access to these insights that the company brain gives them. So I think it’s a very, very different. And so I think in that sense, the humans aren’t necessary.
We’re also not, because we don’t profit off of humans spending more time than is necessary, we’re not trying to build a law firm to compete with traditional law firms. I think in some sense, the goal is that this law firm becomes unnecessary over a period of time because ideally we spend as little time as possible from our own margin perspective. But for this interim period, I think there is going to be a, it’s just unrealistic to expect the transition to go from almost no adoption of AI to AGI.
which is almost how I think a lot of companies are behaving. And so we provide this ramp, I would say. And I actually really believe that, like, again, I’m obviously optimistic, but personally, I believe there’s going to, for every hundred jobs that are being replaced, which I think will happen with AI, I believe there’s going to be a hundred new jobs that are created. And the worst thing that would happen is that the hundred people that are losing their jobs aren’t actually being equipped for the hundred new jobs that are emerging. So I think we’re a little bit providing, like,
Somewhat inadvertently, we’re now building this AI augmented labor pool and training these people to really reinvent themselves. And a lot of them are really eager and desperate for this, but they don’t really know where to go. Law schools don’t teach it. No sort of legal tech company is doing it exactly in this way. So I think there is a really bright future. And I believe that we’re positioning ourselves to be offering something much broader than just software.
Richard Tromans (10:19.686)
Yeah, and I like to point about the, know, the individual lawyers can do much more with the AI. You know, it’s not so much about just, you know, replacement or whatever. It’s, although we may get there at some point in certain streams of work. But I mean, you know, so like, you know, with that team you’ve got with Johnson and Hanna, now outhouse, and I guess others who you may have hired, you know, since you launched, that’s a significant body of human expertise. Then you add the AI bit, then you add the…
aspects that are coming from the clients. I mean, that’s quite a formidable productive force. Think of it as an engine, right? I in terms of your growth, how is, mean, honestly, I know you can’t really get into the numbers, but generally, how is that going?
Omar Haroun (11:07.197)
Yeah, I mean, we’ve had a thousand percent growth this year so far. So it’s been quite significant. And a lot of that we actually weren’t, even though we did come out of stealth a few months ago, that was mainly for hiring purposes because we found there were some candidates who really appreciated the external validation of a big venture around from a good investor, all of that. But we were still fairly quiet about what we’re actually doing up until last month at the summit that we had in New York.
Omar Haroun (11:33.513)
where we really unveiled a lot more and our website now has a lot more detail around the offerings. We’re still not sharing everything, but I think we had a real inflection point where I think this model of, I mean, every customer that I’ve talked to gets a lot of comfort from knowing that they don’t need to make this extreme decision of sort of trusting AI completely or continuing to really have a team that’s not adopting it in any significant way. And this idea of just
I’ve mentioned intelligence, where you’re getting your team trained. You can talk to a human, but obviously the one thing that does change for us to have a successful tech company here is the unit of value shifts from the number of hours or even the number of humans to either something output-based, the number of contracts, right? Or more of a fixed fee model. And at that point, we’re very transparent about where the human is in the loop, where the human’s involved. But the customer also understands that for us to get the kind of margins that we want in the long run.
we’re actually trying to automate as much as possible. So we’re not just giving.
Richard Tromans (12:33.55)
What kind of metrics are you using for charges? mean, are you saying like we’ll do 10,000 contracts of this type for X amount of money or I mean, how does that break down?
Omar Haroun (12:46.471)
Yeah. So I mean, I think like the, the nice thing about going after, I’ll just call it the labor budgets, right. Is that the ROI is extremely clear and tangible. And we’re often seeing five to 10 X kind of ROI on a customer using us because a lot of times it’ll be, you know, this is no secret that even Johnson and Hannah’s team is already acting as an extension of the in-house teams of Stripe open AI, et cetera. And so.
Omar Haroun (13:16.539)
when they can now see, okay, one of these lawyers on my team who’s now fully loaded and trained on the Eudia platform, either is or is not getting through the contracts five times faster, 10 times faster at the same level of accuracy. And if they are, that starts to have some interesting implications. Like if they were previously using a team of 20 to do the same amount of work, they can now have arguably 10 or five people kind of doing that work. And so that’s where the
The ROI is like extremely tangible and it might seem like, okay, we’re actually just reducing the number of jobs. But I think what I found really interesting is none of our customers have like, once they see that this is possible, if anything, they’re even hiring more of these kinds of people because they’re not just doing the old work faster. They’re actually starting to, for example, think about the so what of the red line instead of just redlining. And that might mean, Hey, I actually noticed we’ve never won this.
position in the last 10 deals that we tried to negotiate it, I recommend that we just concede it now and pull forward this revenue much faster, right? And that’s the kind of insight that the general counsel would love to have lawyers who are thinking like that. But historically, we’ve all been so bogged down by just high volumes of work that there’s not time in the day to even think about that kind of thing.
Richard Tromans (14:27.976)
Yeah.
So just to clarify that a little bit. So is it a volume based price or do you just sit down and negotiate and they give you, you charge a fixed fee per year and you just say, look, it’s gonna be X and we’ll figure it out.
Omar Haroun (14:42.567)
Yeah, we’ve optimized for simplicity today. And so it’s typically just a, low flat fee annual fee. And that’s, that’s based off of you typically having proven in some kind of a pilot already. Here’s the clear ROI. And so this is going to be a small fraction of sort of, the amount of value that we’re delivering.
Richard Tromans (15:03.772)
Gotcha. And on the tech side, I know you don’t want to explain everything there, understandably, but obviously you’re using some degree of generative AI. I presume you’re tapping the various models, you’re using the agentic aspects where they make a good fit. mean, anything you’d say more about that?
Omar Haroun (15:22.205)
Yeah, 100%. So I mean, think on the AI side, exactly like you said, I think from day one, we’ve been on the one hand, model agnostic, on the other hand, we have a whole research team constantly figuring out where is Anthropic better or worse than OpenAI and how do we kind of combine the models in a way that makes sense. I think the part that’s honestly the most unique and from day one, Ashish, my co-founder, he built Search at Apple. We have a lot of engineers who have more of a background in
Richard Tromans (15:28.632)
Thanks.
Omar Haroun (15:51.071)
Knowledge graphs and building knowledge platforms. And my own view from the last decade of working in this space is whenever I work with an in-house team, it’s very clear that the person who’s been there for 15 years, even if they have an IQ of 100 is way more valuable than the Yale law grad with an IQ of 160 has been there for one day. And so in other words, it’s not about intelligence. It’s actually about knowledge. And so a lot of our platform has been, and even our engineering team has been all about how can you
really capture and codify and extract this institutional knowledge, which is distributed across a variety of SaaS applications and locking people’s heads. And so a lot of our approach has been really learning over the last two years deeply, how do you do that effectively and efficiently? But once we actually can build this brain, which is more of a knowledge graph than it is an AI problem, then when you apply AI to that sort of, I mean, everyone’s talking about context engineering now. So that’s really where it starts to add a lot of value for the company.
Richard Tromans (16:49.769)
And also preference engineering as well, to some degree, giving them what they want in the language they want and so forth. No, sorry, please go on. You were going to say.
Omar Haroun (16:53.107)
Exactly.
100%. And, and yeah, yeah, sorry. Good. well, yeah, just, just, just kind of to that end. think the, nice thing about having worked with almost exclusively chief legal officers so far is to your point for the, parts that are subjective or where it’s more of a preference, like a stylistic preference. also have the ability to just ask the CLO when it comes to your M&A deals in Brazil, whose lawyer, which of the lawyers that you work with do you want us to model this off of? And even though it may not be objectively right or wrong, I think that’s then allowed the customer to feel like they’re getting the kind of outputs that they really love because it matches whoever the favorite human expert is.
Richard Tromans (17:43.976)
Yeah, and also one of the aspects of knowledge obviously is that it’s a accretive, it’s a cumulative, which of course goes back to your original point, which is that so much in the traditional lawyer, law firm, then client relationship, it just gets lost in the mix. I mean, know, know, knowledge management teams try very hard to capture it, but it kind of gets lost to the client. Interesting, interesting. So a thousand percent growth this year, and we’re not even out of this year, how far? More M&A?
Omar Haroun (18:17.583)
yeah, you know, we’re not as I like, think one maybe miscarriage, but just because we’re probably one of the few series that companies who has now already made two acquisitions. think people put us into this kind of. Bucket maybe of we’re just going to go buy it by a bunch of companies. Yeah. But, but, but, but actually like one of my potentially controversial views is that a very small percentage of people are actually responsible for a hugely outsized amount of kind of progress. I’ll just call it. And so we’ve actually, met hundreds of acquisition targets. And what we’ve concluded is there’s like less than 1 % that actually have these 10x lawyer type of equivalence, right? Of people who are like, even if, whether, whether you call them a law firm or an ALSP or whatever it is, they actually are the best out there. And I think the definition of the best is also changing in an AI first world where historically it was maybe the most credentialed or the number of Ivy league degrees is how you judge the quality of an institution. think now AI is this sort of completely new skillset of really knowing how to leverage it well. And so we actually aren’t on a big buying spree. I think it’s more about where there are companies that meet that bar, which is a very, very high bar. Then we have the capital and we’re going to share more about that over the next few months. But we are in a really good position from a capital raising perspective.
So we just really believe that this entire ecosystem needs to be changed. And it’s not about incremental change. It’s about exponential change. And frankly, kind of once you solve the incentive problem, I’ve always believed everything else will fall into place. But right now the incentives are just unfortunately very perverse for law firms, for lot of tech companies. I mean, just across the board, they’re not in the right direction. And so that’s what we’re trying to really kind of solve here.
Richard Tromans (20:11.484)
Fantastic. Thanks.
Discover more from Artificial Lawyer
Subscribe to get the latest posts sent to your email.