Eudia is a market disruptor that combines an AI company, an ALSP, and a law firm, all designed to help inhouse teams save money and get better outcomes. AL spoke to CEO, Omar Haroun, about where they are now and his hopes for 2026.
First, he notes that the NewModel legal business reached over $20m ARR in December, double the startup’s goal after initially focusing on a group of Fortune 500 clients. Naturally, the plan is to expand on that this year – rapidly.
Another key point he makes is that when it comes to legal AI, it’s a challenge to build for both inhouse and law firms. Or as he explains: ‘Our core belief is that legal tech is not even really a cohesive term because the needs of inhouse teams are not only different, but often at odds with the needs of law firms.’
The ambition is clear: to really make an impact by meeting distinct inhouse needs.
‘We’ve built a data platform, a knowledge platform, and an AI platform for inhouse legal teams. This isn’t just about some low hanging fruit productivity gains for doing the work incrementally better. It’s really about exponentially changing how the work gets done,’ he adds.
Eudia also does not shy away from ROI measurements and KPIs, in fact, Haroun underlines that proving value and cutting inhouse legal spend are core goals for Eudia.
‘The types of outcomes that we can drive for our customers are reducing outside counsel bills by 20%, or eliminating 90% of contract review – really getting to hard metrics,’ he adds.
In short, this is the type of NewMod approach that will – in time – have an increasing impact on the market. The only question is how far and how fast can Eudia and its peers grow?
Press Play to listen to the interview. You can also check out the AL TV Channel for 200 other videos for free.
AI Transcript
Richard Tromans (00:00.215)
Hey everybody. We have with us Omar from Eudia who we’ve spoken to before, but today we’re doing a special interview. We’re gonna pick over some of the key points that are gonna be really coming to life in 2026.
Omar, thank you for being on the show. So let’s talk about where you are now and then let’s get into the future. So first of all, where is Eudia today? January, 2026.
Omar Haroun (01:11.596)
Yeah. So we have, exceeded all of our revenue projections. kind of, we thought we’d be at originally 10 million, ARR we crossed 20 in December and it’s been explosive growth with again, from, day one, our core belief is that legal tech is not even really a cohesive term because the needs of in-house teams are not only different, but often at odds with the needs of law firms. And so we’ve exclusively been focused on the in-house team. And really our biggest milestone, I would say, is that we’ve been very, very focused on driving measurable business outcomes for the chief legal officers and their teams that we help. So that’s a lot of our progress and we’re really excited about 2026.
Richard Tromans (02:01.327)
Yeah, I mean, it’s a double AAR. I mean, that’s pretty amazing. I mean, what was that down to? Just a couple of big unexpected clients or?
Omar Haroun (02:11.182)
Uh, well, so this is like the, the, year, like at the start of the year, we were, uh, you know, uh, a little over two, uh, of, of the, start of 2025. And so we thought five X thing would be a great year. turned out to be a 10 X year. And, um, you know, I, I think a lot of it just came down to. We were very, very kind of, we, we went very slow before going fast, I guess I could say. So we picked initially just 10 Fortune 500 customers and went really, really deep to try to uncover and discover.
what we could uniquely provide that they were desperate for. And it really meant resisting the urge to scale for the first few months of the year. But once we really started to see consistent needs across the board, we then were in a good position to effectively launch in September.
Richard Tromans (02:56.975)
of 2025. And just to remind the viewers, people, I’m sure most people have heard of Eudia now, because the name has rung out, but I think a lot of people still are a little bit fuzzy on what you actually do. Can you just set out what does Eudia do?
Omar Haroun (03:13.998)
Yeah, definitely. So we’ve built a data platform, knowledge platform, and AI platform for in-house legal teams of really skewing towards not just Fortune 500, but what we call some of the world’s most important organizations, companies where this isn’t just about some low hanging fruit productivity gains for doing the work incrementally better. It’s really about exponentially changing how the work gets done and driving measurable business outcomes. So.
You’ll hear me talk a lot about outcomes, increasingly because that, really is like examples of the types of outcomes that we can drive for our customers are reducing my outside council bill by 20%, eliminating 90 % of contract review, really getting a hard metrics. And then what’s made our model a little bit different is, unlike other tech companies, we actually have the human component, whether that’s a forward deployed engineer or a forward deployed lawyer. So we’re able to actually.
offer that outcome end to end and take on some of the risk frankly of if AI is not perfect, then my investors won’t like me as much because our gross margin is going to suffer. But ultimately, we actually can offer in some cases a complete replacement to work that’s going externally or if it’s going internally, an opportunity to actually ensure that the end to end outcome is achieved.
Richard Tromans (04:32.527)
Gotcha. And the people who are your customers specifically, they’re GCs, they’re of legal ops, they’re somewhat, there are other people in the organization who say to the in-house team, you’ve got to use Eudia, you’re recommended by law firms. I guess not, I mean, how does it, who are, who’s coming into your office and talking to you?
Omar Haroun (04:54.028)
Yeah. So we’ve like, I think I’ve personally met over 350 chief legal officers in the past 18 months. so the chief legal officer is in almost every case, the main champion and person who brings EUDIA in obviously at their level, they’re not involved like every day with every, pilot that we’re running or deployment that we’re, engaging on. so, it’s, it’s, it’s definitely been also the head of legal ops, a variety of deputy general councils.
And then the most interesting trend I would say, and since this is being listened to in January, we will have had this LinkedIn live event last month with Gary Hood, who’s a good example of someone who was the chief legal officer of Duracell and after bringing Eudia got promoted to be the head of AI for the company. But we are seeing a lot of CEO, CFO level engagement as well, because ultimately what we’re really building is a blueprint for how do you actually take knowledge work and service heavy functions and transform them with AI.
Richard Tromans (05:54.587)
So on that point, I mean, would you say that you’re helping corporates do work that they would probably always keep inside or maybe they would send out to an ALSP or whatever it was, or maybe they would have to build their own contract management team or some other internal team. It’s that work, you’re not getting into the work where a corporate would say, no, this has got to go to blah, blah in New York because we always send it to blah, blah in New York. Could you just explain which bit are you doing?
Omar Haroun (06:23.928)
Yeah, definitely. Great. Great question. So I mean, I think when you think about knowledge work, it really, to me, raises kind of the question of knowledge and the different types of knowledge. And we’ve kind of obsessed over this internally because again, my co-founder, Ashish, built a search at Apple and was a founding engineer at Google. We actually have more engineers, I would say, who have a background building knowledge graphs and data platforms than, not more than AI, but I would say that that’s probably where we have the top 1 % of the talent out there is on the data and knowledge side of things.
And I, and I believe that knowledge can be divided into knowledge. That’s rare and knowledge that’s commoditized. And so when you talk about maybe like the legal work that’s historically either gone to the, white shoe partner in New York versus kind of the ALSP, I think that was our before AI sort of our, way of thinking about, okay, if the knowledge work is of a certain level of value or risk, I’ll send it to the really expensive human. if it’s perceived to be lower value, I’ll send it to the.
lower cost human. think what we’re seeing and what we believe is that now AI is actually really impacting both categories of work slightly differently in that on the, I’ll just call it like the expert knowledge side or the rare knowledge side. There’s now this incredible opportunity to actually scale the judgment and expert knowledge with AI. And it’s a very, different model than just, I send it to you on a one-off basis, but more.
If I send it to one of the experts at Eudia or even one of my current experts, but Eudia platform is embedded. Can we actually now essentially download the brain of that person, start to really understand how the expert goes about generating knowledge and harnessing that knowledge and finding a way to apply that to the platform on the more commodity work. think that’s, that’s more just completely sort of getting automated and displaced and yeah, exactly.
Richard Tromans (08:16.335)
It sells itself, doesn’t it, that one there? So I mean, in terms of, because you just, again, for the listeners who haven’t picked up on this, you bought Johnson Hanna, which is a well-established ALSP based in Ireland, operates globally, works in…
You’ve or formed a law firm in Arizona, which again, people in the US will know this already is a special jurisdiction where you’re allowed to create alternative business structures. Just talk us through what is that structure for? Is that effectively almost like just like a nameplate that allows you to then do certain things? Cause you’ve got a license there. Do you have like a whole room full of like very experienced lawyers in there?
I mean, what’s that?
Omar Haroun (09:03.404)
Yeah, definitely. Well, I think, I think going back to kind of this, this differentiated types of knowledge, what, what we found is that the people who are at the top of their field are, not only like a little bit better, but exponentially better when you combine that knowledge with AI. so our, our, our approach, I mean, I would say there’s really two main benefits of having the law firm and sort of this kind of services, side to the business.
One is that just in practice, it’s actually much easier for an in-house legal team at times to engage with a law firm than it is to engage with the technology company. Historically, they haven’t really bought a lot of technology. They’re not really set up to do that. And so when you’re selling software to an in-house legal team, there’s just a very, very different kind of set of cycles that you have to go through. So one is literally just how do we make it easier for our customers to consume our technology and actually being a law firm can help with that. So that’s one very practical benefit.
Richard Tromans (09:59.951)
Actually, just on a technical point, I’m not sure if it works the same way in America, but in the UK, generally a law firm, you have to send effectively a client engagement letter with a letter head and we are regulated by blah, blah. And the client gets one of those and then you set out terms and then off you go. So this is what you’re doing now.
Omar Haroun (10:18.06)
Yeah, mean, for, for customers who engage our law firm, there’s an engagement letter. And to be fair, there’s a different process if it’s the law firm versus the, the sort of, formerly known as Johnson and now you, you do Europe operation. But, but I think the broader point is, is buying services for the chief legal officer has generally been much easier than buying software. And the reality is AI kind of blurs the line. You could, you could portray it as either, right? Because ultimately.
to really get the value out of it. You do need some combination of humans and AI. And so are you, are you going to say that it’s a service provider or a software company? What we’ve learned is that it’s actually often easier for our customer to procure us if we’re perceived as a services company, but we have no desire to like, we would never even allow the unit of value to be the number of hours or the number of humans. It needs to be something output or outcome based. So one, one benefit has just been, it’s just make it easier for our customers to consume the technology.
And then the second benefit I would say is like, look at some of these workflows, particularly the ones that go to outside counsel today. And, we’re not saying that our MNA due diligence tool is, you know, 10 X better than the other MNA due diligence tools out there. I mean, we, we, we happen to believe that, but that’s not the pitch. The pitch is, Hey, meet one of our MNA lawyers. I just talked to this guy who practiced at Gibson Dunn for a number of years, very, very solid attorney, the kind of person that most of our customers would pay.
$1,000 an hour to, do their &A, diligence work and historically need 30 of those people to really get a deal done. What we say is let us show you what that person paired with our platform can offer. And they get a lot of comfort from seeing that. And frankly, also from a quality standpoint, the quality is just much higher when you have the human in the loop and not just any human, but actually a highly trained human who knows exactly how the big law firm would, would do the work and then finds a way to just.
have the incentives fully aligned with the customer to get back issues lists in five minutes instead of, you know, five days and really offer significant benefits.
Richard Tromans (12:19.439)
So the model going forward really is probably increasingly so it’s going to be you have people who are big law partner level inside Eudia and you have it you have the tech you have workflows the data the knowledge you have the ALSP bit if you need that extra leverage on the human side you’ve got the whole shebang really.
Omar Haroun (12:37.71)
Yeah. And, you know, I mean, it is interesting how we are starting to work with law firms as well, because actually for M &A and litigation, which are probably the two most common examples where a customer does currently outsource the work to law firms in a significant way, you’d actually been surprised how receptive, at least like the top law firms and the very forward thinking individual partners have been. Part of why we haven’t sold to law firms is that law firms are very, very complex organizations. And I actually just perfectly don’t like partnerships as a structure.
I think it’s very, hard for anyone to make decisions. so what’s been interesting is now that we’ve been around for a couple of years and I think a lot of the top law firms have seen the reality is if it’s a high stakes litigation matter, they’re going to get plenty of work no matter what. So they’re actually now using our software as well. So it’s not only that we have a model where we provide, where we insist use our law firm or nothing.
Richard Tromans (13:26.761)
So you’re getting revenue from low firms now as well.
Omar Haroun (13:30.222)
Um, we’re not getting revenue from law firms, but we are having law firms use our platform. And that’s typically initially at the direction of the customer. if they already have it. Exactly. But ultimately I think what’s, what’s been interesting is how open the law firms have been to that, because I think that they’re starting to realize AI is not actually coming for their job. It’s just changing the work. And so whether they’re our platform or different one, they typically don’t try to push their own platform on a customer who has no desire to use that platform.
And I think the best lawyers are starting to realize they may not be experts in AI, but they are experts in learning how to use AI.
Richard Tromans (14:06.195)
Interesting that when you say is, know, dispute litigation, mean, you’re doing e-discovery work for them, you’re doing other types of things. So you can, weigh beyond the contract side now.
Omar Haroun (14:12.894)
on the, yeah, on the, on the litigation side, we’ve, we’ve mostly, today helped with, really high stakes litigation and a lot of it’s been at the pre-trial or trial phase where, because our platform’s already accumulated a lot of knowledge about this customer, we can very, quickly ingest a lot of data, for, for things like being able to flag inconsistencies and what
you know, kind of some of the expert witnesses or people even who are part of the trial have said, really, really kind of going to that level where it’s not the discovery work as much as it is the, really, really next level. How do we think about the outcome being winning the litigation that we can help drive in that, that example? So that, that’s been one area that we spend a lot of time with litigation, but, obviously for a customer who’s already had Eudia build this brain for them, where we’ve ingested a lot of knowledge and data about the customer. There’s also some ways to even just for, for, basic.
search question and answering, really kind of taking all their data and making it more accessible for the litigators.
Richard Tromans (15:14.063)
Okay, let’s talk about the brain. Let’s talk about the brain. So how much data, you know, in an optimum, optimal situation, would you like them to give you? mean, mean, data storage now is virtually unlimited. So I imagine they could give you pretty much as long as it’s digitized in a sort of sensible way, you can handle it. But what do you want from them? You want all of the contracts for the last 10 years? want, what are they giving you?
Omar Haroun (15:40.63)
Yeah. Yeah. So I think one, one big thing for us this year that you’re going to hear a lot more about over the next few weeks is partnerships. And I think we’ve started to appreciate that. our customers have invested a lot in their existing technology stack. So whether that’s again, if they’re Microsoft, Amazon, Google shop, whether they’re using service now Salesforce or a CLM, we’re, we’re not, you know,
building everything ourselves, we ultimately see the brain as this intelligence layer and also unified data platform. so one thing that we’ve done differently from day one is actually, again, like I think the law firm got more attention because that was more newsworthy, but a lot of other people are starting to kind of create forward deployed engineering teams. That’s been our model from day one where we believe every AI problem is a data problem. so, you know, kind of a lot of the work has actually been
already connecting to a lot of those data sources internally, because once you go beyond the pilot, it really becomes critical that you actually do have a way to connect.
Richard Tromans (16:42.663)
Well, and this is another way of looking at it obviously is that they don’t actually have to, nothing has to leave their building as it were. You go to them and they tell you, they open up their kimono to varying degrees and saying, right, you can have this bit, you can have this bit, you can have this bit right now, come back and bring some value to us.
Omar Haroun (16:49.538)
Exactly.
Omar Haroun (17:00.298)
Exactly. And I think the part that’s been really exciting is to your question, we actually don’t need like to prove the value. We don’t need that much data. we could have a hundred contracts along with the requisite financial and operational data. And that might be enough for us to be able to tell them things like, like, if like, like one, one customer, the outcome that they were very focused on us delivering was around gross margin improvement. And just that limited amount of data was enough for us to be able to tell them, did you know that,
When you use this language in your contracts, your margin is actually 50 basis points higher. And, know, it may not be a conclusion that’s justified when you look at the full set of data, but even just kind of helping them demonstrate that we’re actually able to take these data silos and actually. Built it’s kind of set of actionable insights. That’s kind of along the lines of the outcome. They actually want to drive is something that’s been easy to do with a limited amount of data.
Richard Tromans (17:52.737)
Interesting, interesting. mean, I’ve got to say the word moneyball. mean, there’s a degree of moneyball going on here, isn’t there? You’re taking that data, you’re analyzing it, and then you’re playing it back.
Omar Haroun (18:06.636)
Yeah. And that’s just one example where the outcome that we delivered was around gross margin improvements. And so the contracts and the financial data was the right way to achieve that. But when you look at something like reducing my outside council bill by 20%, that’s an outcome that we can also achieve. But obviously it requires first understanding where the spend goes. If it’s a customer who’s, whether it’s &A litigation or even just their current work, I think that’s a very, different path. It’s less money ball-ish and it’s more…
actually just taking over and starting to automate a lot of the tasks in those workflows.
Richard Tromans (18:34.809)
down.
Richard Tromans (18:39.279)
I mean, just out of interest, and I guess there’s a question I’m going to put to every new mod that I meet over the year, over this year, is how far up do you want to go with the value chain? mean, presumably there are certain points where you’re just like, no. And the client, these are sophisticated clients. They’re going to be like, look, we’re about to buy a $55 billion company. We’re not going to use the primary advice. Just not going to happen, right? But we’ll use you up to here. And then White Shoe Law firm in New York takes over.
So you better get old together. I mean, how far are you going up?
Omar Haroun (19:12.558)
Yeah. Well, so I think like originally, honestly, two years ago, I would have said, Hey, we’re only going to take the more low value work. I’ll just call it right. The high volume repetitive. Um, the part that I find really, really interesting is, and I think we may have talked about this a little bit last time, but like, I’ll just take red lining as an example, right? There’s a thousand, I’m sure you’ve met a thousand companies saying they can do red lining. Uh, I was talking to one of, one of our customers recently and he said,
Richard Tromans (19:23.087)
You
Richard Tromans (19:36.757)
Omar Haroun (19:42.186)
He believes redlining is a crutch for people who don’t know how to do deals. And that was a really interesting comment because what he really meant by that is you look at the status quo and it’s basically at the top, at the very, very senior expert levels, you have a lawyer who, like the chief legal officer, who actually has developed incredibly powerful judgment and intuition around how to navigate what I sometimes call the speed risk trade-off, right? Where the life of a chief legal officer is the business wants you to go faster, faster, faster. You obviously need to.
manage risk, but not at the expense of kind of slowing down things too much. and I think the, the, the CLL is not to do that, but what ends up happening today is they create these out of necessity because they have to scale themselves. They create these playbooks and you now have these like rules that get pushed down to people at more junior levels or external people who are lacking context. And each person in the value chain actually is now incentivized to be more risk averse because no one gets fired by being more risk averse.
But you end up with this like incessant redlining and the other side has the same problem. And then actually what ends up happening when you really study how this process works is five weeks later, the CEO gets frustrated. Hey, what’s happening with this deal? And then the CLO or someone senior comes in and says, Hey, it’s fine. We can live with limitation of liability being 10 million, even though that wasn’t one of the rules because they actually understand the principle. And so a lot of what we’re actually doing is shifting, I would say the whole industry from this very kind of.
process heavy rules based approach to a more, outcome heavy principles based approach. And unlike other companies, we’re not trying to help you do your red lines slightly faster. We’re actually trying to get rid of redlining, right? In some sense, because that would actually then unlock the outcome of closing deals and 90 % faster, right? Which is what the business actually wants.
Richard Tromans (21:32.579)
Gotcha. Yeah. So you’re coming at a more sort of higher level strategic. And I guess at that point you can go the sky’s the limit to some degree. Isn’t it really? you, if, if the idea is to really almost like tune into like the meta legal risk issues of that company. above stuff is probably not even written down, but you can perceive it by understanding their data.
Omar Haroun (21:48.961)
Exactly.
Omar Haroun (21:54.668)
And, and, and that’s, that’s kind of my personal view. Like people always talk about data as the moat and, you know, it’s, it’s, think a very oversimplified statement. My view is the undocumented knowledge of experts is actually the moat. And that’s what we’ve been able to acquire, I would say, by really getting very close to these chief legal officers. And when I say download the brain, I actually mean, you know, with, with, with their consent, of course.
really deeply understand how are they actually, how do we go from intuition to data and really start to understand how they think so that we can actually surface the right data to make their intuition even more, less intuitive and more data-driven.
Richard Tromans (22:39.311)
Yeah, yeah, that’s very interesting. Yeah, exactly. So yeah, what’s not documented is the most. Well, let’s look at the future. So it’s 2026. last couple of years have been incredible. I’m sure everybody would agree. This year, no doubt, will be the same, if not more so. There’s a whole bunch of other new model law firms, AI first law firms, people have different terminology. There’s a whole bunch of them now, more popping up every few weeks.
Where do you see this going? mean, someone posted something on LinkedIn a few weeks ago saying, you know, like when will the first AI first new model law firm join the AMLOR 100? And I said, well, it won’t be for a few years yet, because if you look at the total revenue at the bottom of the AMLOR 100, we’re looking at several hundred million dollars. That’s a lot, right? Don’t underestimate how much, long it takes to accumulate that. But I said, it will happen.
It’s just a question of time. There’s no theoretical boundary that they cannot cross, right? So we just wanted to get your view on that. What does it mean? Not just your company, but others who are doing something similar, sometimes in very different spaces. How is all this going to change now?
Omar Haroun (23:53.964)
Definitely. Yeah. I mean, I think my view is, there is a, like, you, you know, as well as anyone, this has been an extremely slow industry to change. think all of the ingredients are there now, for it to, to not just change a little bit, but to fundamentally change. But fundamental change is not that it just can’t be that fast when you have these extremely entrenched institutions and incentives that, and lobbies that are very, powerful.
So think my view is probably more along the lines of like the next three years is where I see something like that happening because what I think is going to be the path is first of all, people have to actually see this is really not only possible, but it’s actually proven repeatedly that like, think ultimately the one challenge with a lot of the legal AI companies who sell to law firms is on some level, they now have a conflict of interest if they also sell to in-house because at the end of the day,
Um, if, they’re, if, if their primary job is to actually help law firms preserve their margin or which, which a lot of them have said publicly is actually what they’re working on, then that’s not really what the in-house team wants or needs, right? They actually need a lot of this work, internal and external, that can be automated to be automated as quickly as possible. I mean, ultimately their, their, their job as a chief legal officer is to actually drive earnings per share for the company, just like every other corporate officer in the company. And so, um,
Richard Tromans (25:00.195)
and
Omar Haroun (25:19.53)
They’re not here to help train the lawyers of the future. They’re not here to help law firms preserve their margins. so, you know, I think part of why we frankly not seeing the kind of speed that we probably ideally would see is because there’s a lot of people who are trying to somehow make sure that the current model is maintained in some way or it’s not too disruptive. But I actually think there are people like us that are out there and now several others cropping up who don’t have that problem and they are unabashedly doing what’s best for the customer, which is the kind of end corporation.
Richard Tromans (25:47.751)
And it has to be said, it’s, mean, only, only in the legal, legal tech world would we have a debate like this. you know, quite often the, the clients, it’s like, yeah, there wouldn’t be a legal market without the client. We’re so used to talking about, you know, what, what the law firms do and what the law firms want and how much profit will that partner make this year? And what’s the associate bonus going to look like? like, yeah, but you know, none of that exists without the clients, right? You know, law firms exist by themselves.
Omar Haroun (26:01.121)
Exactly.
Omar Haroun (26:14.07)
Exactly. Yeah. And so, so, I mean, think there’s, there’s, actually kind of like, two steps and we’re at the beginning of proving step one, but I think step one is proving this actually works. And again, I think, I think our model of getting some of the world’s best lawyers combined with this platform so that you’re not just accepting this is like the low value work. actually is. like over time, the high value work as well, is, part one, but I think part two, which is that, which I’m really excited about is we’re seeing individual partners.
Who are even at the current law firms and plan on staying at the current law firms. I think one nice thing about the law firm model today is because it’s a fiefdom and every partner is out looking for their own book of business. You actually do have opportunities that we’re not starting to see where individual partners who actually want to put the client’s interests first are partnering with us and companies like us. And I think what’s really interesting about that is even if they take a hit, like, like let’s say one of our customers does 20 &A deals a year, they now plant UTIA.
for all the due diligence across all 20. We either need to just hire more and more of our own M &A lawyers, but honestly, that’s not really our model to scale. Or we find out of the five law firms they use today, they can basically choose the one or two where the partner gets on board with this model. And even though they’re foregoing 50 % of the revenue they would have gotten on each deal, they may now get 10 deals instead of two deals, right? And so it actually starts to become, I think once the law firms begin competing,
and realize that actually some version of the fixed fee model adopting the technology the client actually wants to adopt will actually help them win more of the client’s business. That’s when you’re going to see the real sea change in my view.
Richard Tromans (27:53.871)
Or does one of these large law firms say, right, we’ll buy you, People have suggested that to me, but I mean, you’d be like, well, I presume you’d say, no, thanks, we’re scaling.
Omar Haroun (27:59.694)
Yeah.
Omar Haroun (28:09.728)
Yeah. I mean, I look, I think there’s, just going to be like what a law firm even means is going through such a big change that, it would take a pretty radically different law firm and one who probably is not a partnership. I think that’s, that’s really the uncomfortable truth is that when you have a partnership and you have to have some version of unanimous or majority consensus decision-making, I think there’s just enough people who are five years away from retirement that
Richard Tromans (28:15.928)
you
Omar Haroun (28:34.498)
they have no desire to actually fundamentally change when they’re, when they’re buying their vacation home that they want to buy with the current model. Right. So I don’t see it really happening unless there’s a firm that actually can really move at that kind of speed and consensus, which seems very, very unlikely.
Richard Tromans (28:49.455)
And they’d have to disrupt themselves quite considerably, quite considerably. So how does this pan out then? I mean, you know, the, do you think the clients are changing? I mean, again, you know, it always does come back to the clients. I mean, you, you you just said earlier, you spoke, spoken to literally hundreds of senior in house lawyers, CLOs, GCs, et cetera. Do you feel that they are actually changing their worldview now?
Omar Haroun (29:15.566)
100%. And I think there’s the level of like, what I, what I really believe is we’re six months away from this, this sort of, um, come to Jesus moment where I think everyone who’s been investing and almost in a frenzied way in 2025 and the first half of 26 frantically investing in AI because they want to prove to their CEO that they’re actually AI first in some way without really thinking about why or what are the outcomes I’m going to drive.
are going to be in a very tough spot because these tools aren’t cheap. so it is a real investment. And if there’s no real tangible ROI or outcomes that you can point to, I think you’re going to be in a very, very tough position. And that’s, that’s where I, I see on the corporate side. I mean, the good news is every other corporate officer is facing the same pressure or sees the same opportunity depending on their, on their outlook. But when you’re a chief legal officer and you’re seeing your
head of HR, head of finance, head of supply chain, all bringing really, really innovative ways to drive value. And I think many companies actually do have targets, right? They have hard, whether it’s related to headcount, cost reduction, or just other value propositions, it is a mandate at a corporate level. But I think right now a lot of the legal tech is more like a toy that actually adds some kind of productivity and everyone likes it. But it’s very, hard to pinpoint where the actual ROI is.
Richard Tromans (30:38.551)
Yeah. mean, know, I mean, it’s a subject I spend a lot of time thinking about, too much time. But I mean, one of the things I have to consider is that it’s because people don’t have expectations that legal can change significantly that these things don’t get taken so seriously that once you actually, it’s, know, it’s like, you know, if you’ve got like a young kid, you don’t expect them to be able to drive, right? Or fly a plane or so because their expectations are not there. You don’t, you don’t.
measure what they’re up to. You’re just having a golf and enjoy yourself, have fun. You’re a little firm, enjoy yourself. But when you start to change your expectations, you’re like, well, really, so I can lay down measurements the same way that I do with HR on the legal team. Really? Oh, I didn’t realize I could do that. Oh, great. Well, that’s good. Let’s start doing that. And of course, once they start doing that, it just builds this new culture.
Omar Haroun (31:36.47)
Exactly. Yeah. And I think, I think that this is where, I mean, I already mentioned Gary Hood, you know, is now chief business and AI officer for the company. There’s probably about a dozen of our customers that are just now at the point where they can actually share. Here’s the $25 million of hard value that I was able to deliver. And I think once that starts happening and now that becomes the expectation as opposed to right now where it’s kind of,
Richard Tromans (31:40.569)
Hmm.
Omar Haroun (32:03.918)
My legal team really likes this tool and that’s sort of the change that I’ve made, which I don’t mean to dismiss it because it is actually a big milestone to get lawyers to go from not using technology at all to using it in some way. But I think we’re now quickly approaching the point where if there’s not also some kind of ROI, it may not be clear why you should spend $500 a month per license for some jewel that you roll out to your whole organization.
Richard Tromans (32:29.103)
No, totally. So what you’re saying, I’ve had this argument back and forth with various people over the years online. What you say to the person who comes in and shows you the chart that says, well, legal spending is between 0.5 and 0.02 % of revenue at these corporates. They don’t care. Stop talking about efficiency. It doesn’t matter. Just let them do what they want. How would you answer that?
Omar Haroun (32:57.186)
Yeah. I mean, I think, I think it’s part of why we’ve sort of as an industry, been able to avoid a lot of change today because the combination of that and just this market being highly nuanced for anyone else to figure out how to sell into just, just means legal has been able to live on this Island, right. And mostly left alone. my, my, my own view is like two things are changing. One legal just happens to literally be, I mean, even if you just ask any person who’s
learning about AI or in a computer science department, pick your top three industries to go into where language models can have the biggest impact. They would all say legal, right? Just because it actually is an industry of language and words and, language generation and language analysis, basically is what lawyers do. So I think there is just like an element of, we probably have reached the tipping point of just when there’s like a thousand legal tech companies where there was, you know, 50 before, um, it, starts to get, to kind of become very, hard to ignore. I think the other part is, and this is kind of,
Like borderline more humorous or anecdotal, but I haven’t met a single CEO who doesn’t actually get like quite annoyed by the fact that they’ve all dealt with usually law firms who seem to be just billing a lot for work that they now know from their own private use of chat. GBT can be automated. so like one thing I’ve actually noticed when I talked to the CEO is there is an element of like.
It’s just frustrating and it may not be like, it’s a huge line item for them to be fair, but if there actually is a way to really just like not have your dollars go to paying this tax where people are basically doing work, manually that we all know can now, can now be automated. They’re actually very on board. And I think the best chief legal officer see the same opportunity, which is it starts to almost become a dereliction of duty. If you actually know the work can be automated, but you’re paying kind of parades that indicate that it’s.
Richard Tromans (34:52.679)
Well, you stepped on a hot potato there, which is, are you reneging on your fiduciary duties as a director or manager of a business not to try and save money when you know it can be?
Omar Haroun (35:06.35)
Exactly. And I actually believe, I mean, I wrote a paper about this a years ago, but I actually believe that it’s unethical not to use AI, right? People always talk about ethics and AI and usually in the context of the very justified challenges in terms of bias and all that. But I think there’s also this other argument, which is exactly what said.
Richard Tromans (35:25.871)
Yeah, yeah, yeah. Really interesting. Really interesting. Well, just to wrap up, as mentioned, it’s been an incredible journey for you, for a lot of other people the last couple of years. Just looking ahead very, very broadly, what you think we’re going to see unfold across this year, just very broadly.
Omar Haroun (35:43.084)
Yeah. I look, I think I have two, two main predictions. one is that one, one I already more or less said, which is I think the second half of the year is going to be the year of show me the ROI and really kind of the outcomes that you’re unlocking, not just the toys that you’ve been playing with. and I think it will take like honestly, six months before we get to that point. So the first six months will still be a race to get the shiniest toy, but, but then I think it’ll start to become,
wait, how much does that toy cost and what is it doing for you? I think the second one is I actually think there’s going to be a little bit of a recognition in the tech world more broadly. And I hope this is true of the legal tech world, that none of us are going to be able to build everything on our own. so whether that means more consolidation, whether that means just people being more clear about their swim lanes, you know, like when I look at the enterprise stack overall, I look at companies like ServiceNow, Salesforce.
Richard Tromans (36:13.375)
There.
Omar Haroun (36:40.404)
Who have built incredible technologies, right? I think, historically, I think everyone got a little bit carried away with like over the last 18 months, at least of wow, AI is here. We can now all do everything with it, but there may be more of a recognition that, that actually like someone like ServiceNow, like solving the workflow management problem is like massively hard. And to be able to do that at the level of quality that they’ve done is phenomenal. That, that doesn’t mean that they all suffer through the company who’s figuring out AI and legal.
Richard Tromans (37:05.103)
There.
Omar Haroun (37:09.038)
to drive outcomes and going really deep in every vertical. So think we’re going to start to see more partnerships with technology companies. I hope like within legal, that means actually the CLMs and the eDiscovery platforms and some of the folks who’ve actually built great technologies also begin to recognize like they will have their own AI, but that may not be what they’re natively going to be best at. And so there’s a lot of value in partnering with, with AI native companies there.
Richard Tromans (37:34.903)
Interesting, interesting. Well, as always, to be continued. Thanks so much. It’s always a pleasure. And we look forward to seeing how the company grows throughout the year. And we’ll hear from you very soon, I’m sure. Thank you.
Omar Haroun (37:44.758)
Yeah, likewise. Thank you.
Discover more from Artificial Lawyer
Subscribe to get the latest posts sent to your email.