← Building One with Tomer Cohen
Building One with Tomer Cohen
Building Harvey With Gabe Pereyra: Ethical Walls, Agents, and What AI Can Unlock For Law Firms
00:33:52
Loading summary
Transcript53 lines
- [00:00]
A
If you think of what the best associates do for a partner, it's exactly that. I read the whole data room. Here's all the issues I flagged, here's the things I'm uncertain about. Here's my draft of the memo. And the partner reads that, checks a couple things, and they say, I'm confident enough to now go and give that to the client. Because at the end of the day, this client is paying this partner a lot of money to be able for that partner to say I'm willing to risk my reputation that this transaction is correct. I don't think there is in the short term a magic solve for that through AI. I think it's still the the same problem.
- [00:38]
B
The holy grail of building a great product is when you can combine two things. A deep understanding of the problem and deep expertise on how to solve it. That's the case for today's episode. On one side, you have a founder who deeply understands the domain and how work actually gets done there. And on the other, a founder who understands AI at the frontier. My guest today is Gabe Perriera, a former AI researcher. Together with his co founder, they've built Harvey, one of the leading AI products in legal today. It's used by top law firms and enterprises around the world to draft, analyze and reason for complex legal work contracts and cases and filings. When you think of legal, in many ways it's the perfect domain for AI. It's language heavy, it's logic heavy, and it's high value. But it's also very hard with AI because in legal, every word matters a lot and every mistake can carry massive consequences. So you need precision, accountability and trust to get it right. So what does it actually take to build an AI in that kind of environment? I'm Tomer Coyne and this is building one. Let's get into it. Gabe, it's a pleasure to have you on the show. Thank you so much for joining me.
- [02:03]
A
Thank you so much for having me.
- [02:05]
B
So let me start with a personal journey and yours is always ignorant. All builders journeys are quite unique given where you are right now. But yours is very unique versus our guests. You come from a very academic background. In fact, as far as I know, both your parents have a PhD in computer science. They both worked deeply in AI. In many ways you're continuing the family legacy and you actually started as a scientist doing AI research. And I'm curious was just that, that was in the air at the house. You kind of, you saw where this was headed and you decided you want to go there, you kind of, this was a great, you know, sense of like where you can have impact from your family or something else.
- [02:46]
A
I actually think about this a lot. Like when I was younger I wanted to play soccer professionally and so I dropped out of high school, I went to Brazil, I played, got injured. And then after that.
- [02:55]
B
What's your favorite team?
- [02:57]
A
I'm half, my dad's from Argentina so I'm half Argentinian. So it was like a Barcelona. Yeah, exactly. I was a huge Messi fan growing up and still, still follow them. But after that I actually wanted to do finance, like investment banking, start a hedge fund, something like that. I kind of didn't know what I was doing, but I was like, this seems competitive. I like soccer, that was competitive. And my parents were like, you should try computer science, you would love it. And I was like, ah, I don't want to do what you guys do, got to do something different. I went, I worked at a hedge fund and then I kind of worked at a quant fund and through that found AI when I was in undergrad in college. And I think just through kind of reading about it, this was around 2012, this is when I kind of found deep learning. And this was very early days of a lot of the technology that now I think is working really well. And my parents had done related stuff and so I kind of got more interested and I think like from very. Even when I was doing AI research back then, it was always a means to an end of. I wanted to be very technical and I wanted to understand this stuff. But the goal was always to start a company. I just at the time didn't realize that company would be illegal.
- [04:07]
B
So the goal was let me learn the foundations behind the technology really well. But you always wanted to move towards the company side, the application side of it, the business side of it.
- [04:17]
A
Exactly. I got drawn to it through like in high school when I was playing soccer, I was really interested in just how do I get better at soccer faster. And I kind of read a lot of psychology and I think through all of that just human learning in general was super interesting. And, and then through kind of doing that my parents background, I eventually found machine learning and I was like, oh, this is like. And the more you kind of learned about it even then I think I just had this strong intuition of like we were going to be able to build something like AGI. I think at the time, like no one knew the shape of it. But as you saw these models getting better, you kind of had this intuition that this was coming.
- [04:53]
B
And AGI was a dinner, you know, a family dinner conversation on a regular basis when you were young, we definitely talked about it.
- [05:00]
A
So my mom, she's retired now, but she ran the autocorrect team at Apple, and so she actually had built one of the first billion user language model products. My dad was a math professor at Stanford and a bunch of other places, and he did numerical analysis, but a lot of linear algebra stuff, which I think underpins a lot of these algorithms. And so as I was doing more research, I would just talk to them about it. They had both published papers and knew a bunch of this stuff. So I think that was cool just being able to talk to them about it.
- [05:30]
B
Before we get into actually building Harvey, it sounds like you and your co founder, it was the perfect duo. On paper, it looks amazing, right? You come with amazing background in frontier AI and then your co founder, Winston Weinberg comes with amazing domain expertise in legal. You know, it's a dream come true for companies, dream come true for VCs to invest in. You guys. I'm curious about why you chose Legal, because I'm sure, again, given your intention, I'm going to learn this really well and then I'm going to build a company. I'm sure you saw so many opportunities across.
- [06:04]
A
Yeah, I think the one thing that I looked for when picking a domain was actually, will this have a big societal impact? And so when I was doing AI research, the original startup that I thought I wanted to do was personalized education, because I think for me it was really important. This technology is going to be like
- [06:25]
B
a personalized tutor kind of thing that knows you really well.
- [06:27]
A
And how do you use this very powerful technology to go not just build a big business, but solve like a really meaningful societal problem? And so I was actually pretty agnostic to the application. And then I think Legal had that property too, where we're starting to do a bunch of work of how do we help Court systems and access to justice. And so I think there's something beyond just this is working kind of financially. And then I think the bigger thing, obviously was just Winston and I were super close and he was kind of the perfect person to start the company with. And we actually, when we started thinking about the idea, I was brainstorming with some other friends that had worked at Google Brain of kind of more general ideas. And one day Winston said, you know, why don't you build an assistant for Legal? And showed me his workflows and kind of the work he was doing. I Showed him the models, and then we pitched that to OpenAI and they said, we'll fund this. And that was kind of the initial
- [07:20]
B
spark for it, as AI labs keep kind of moving up the stack and they're potentially taking parts of the application layer. How do you think about it, Harvey, as a vertical AI company specialized in a very valuable domain? Did you think about that risk at the beginning? Was it just like, I know my path, or is it, we're going to figure this out as we go?
- [07:40]
A
I think even when we started the company, like, a lot of what we thought about was like, how do we build something that exactly what you said doesn't get absorbed as the models get better? I think that thinking has evolved a lot of what are the things that are defensible? But I think what I would say and the big shift that we're seeing now is I think people still, for the most part, think of these products as individual user products. And so they think about the problem that you're solving is, if only I could make this model smart enough so this user can do this task that they're trying to do. And I think the real problem that we are trying to solve is how do you help law firms improve their business and how do you help their clients get better legal services? At the end of the day, the reason that I pay a law firm to do a large merger is I want them to be accountable for the outcome, right? And it doesn't matter if the model is smart enough to do that. There's like regulatory insurance, human accountability. There's kind of all these reasons on top of why you need kind of these existing systems. And I think this will change. Like one of the nicest examples that I like that's legal specific for when we get the question of like, oh, why isn't OpenAI or Anthropic going to do everything? The simple answer is just conflicts, right? Like if you're a law firm and you just use only Anthropic's products and now you need to represent Google OpenAI cursor like any company that competes with Anthropic, you can't use their models, right? Because none of those companies are going to let you send their legal data to these models, right? And if you think of as these tech companies get bigger and bigger, they're all going to build their own models. None of them are going to let you send their sensitive legal data to competitors. And so you need to build this infrastructure that abstracts this all away.
- [09:29]
B
The idea of conflicts or kind of subject Matter is unique to legal in many ways and it's not easy to solve.
- [09:38]
A
When you think of these law firms, It's I have 10,000 clients and now I need to do that in a way where all of their data is separate and but my lawyers are able to work across that. And so I think there's problems like that and then there's equivalent problems. When you get into enterprise, we work with a bunch of large banks, private equity firms, insurance companies that all these regulatory challenges that you need to solve to deploy these. And so I think the more that I've done this, the deeper appreciation I've gotten for the problems just any of these enterprise SaaS companies are solving. Where it's like once you get to these massive companies and have to solve their problems, you're like, oh yeah, this is like very difficult.
- [10:13]
B
Did you guys were able to solve the training part as well or mostly the separation of access to information?
- [10:18]
A
What we are starting to do is thinking about how do we help a law firm and a client train a model together. The problem you run into is when you start thinking of all of the data a law firm has a lot of that belongs to their clients. And so you can't take all of that data and train one model because now you're breaking all of these ethical walls and these conflicts. And even if you redact the data, the models are now so good that that they can just look at all of the information, even if it's redacted, and still roughly guess like this is probably this company. And so one big thing we're actually working on right now and hopefully open sourcing soon is I think we've made massive progress in how you build synthetic data sets. And so the ability to basically build these RL environments of a transaction, of a litigation that you can train an agent on. And so you can get a lot of the performance gains from either those synthetic data sets or we use a bunch of contract attorneys and lawyers to improve those.
- [11:19]
B
If you're talking to a founder right now and they're kind of battling where you were maybe several years ago, trying to figure out where to go and what to do, because you were like, this is incredible time to build. There's so many opportunities, but I kind of want to choose my path correctly. Is there a framing in mind that you have to offer kind of future entrepreneurs?
- [11:38]
A
I would say the high level advice is you need to find the things that aren't obvious right now. And so like a lot of how I thought about what company to start when I was in high school, and I was thinking about, okay, I maybe want to start a hedge fund. I read a bunch of Warren Buffett shareholder letters, and a lot of what he talked about was like, you need to be very correct, but you also need everyone else to be wrong about the thing you're correct about. Because if you're very correct, but everyone believes the same thing, there's actually not a huge amount of upside for your company. And so I think back in 2014, what I strongly believed was AGI was coming. And I think most people in the world didn't believe that. Right? And if you had made. I didn't have enough money to make investments at the time, but if I'd made the bet of, like, Nvidia, right, that would have been a great bet, right? And I saw, like, I worked with Geoffrey Hinton, and he was like, we need so many GPUs to train these models, and Nvidia, like, is starting to make these for us. And so there was things like that where if you had this really strong belief back then, and then even when we started Harvey, most people, like ChatGPT didn't exist at the time. And so doing this vertical AI solution wasn't obvious. And then people had this belief of, oh, if you build on top of the models, that won't work for all these reasons. Right? And so there was all these counterintuitive things. I think the really interesting companies starting now are going to be the analogy to companies like Uber, TikTok, Airbnb, where before the Internet, you just couldn't imagine these companies, right? And so what are the companies that you can't imagine right now that are going to be made possible by this technology? And I think the best way to figure those out is you just need to use the technology 24 7.
- [13:20]
B
When I think about the product development life cycle itself, you're going to be blocked on ideas soon because actually producing them. What was like two years ago was like, oh, that's fine. Now it's like, you know, now it's clearly like the, like the way to build. And now you're going to start moving up the stack into other areas.
- [13:39]
A
I would actually say the biggest thing like, we're struggling with is not just the bottlenecked on ideas, but bottlenecked on coordination. Like, I think the thing that is increasingly getting hard now is like the alignment, where now our PMs can generate entire product suites. And they're just like, here's how I would build this entire product line fully mocked out in Our design system. And it's like the hard part is like, okay, you have 10 that all have these. Like, how do you allocate resources? How do you solve all, like the merge conflicts of these?
- [14:09]
B
So when you started, was it, I'm going to pick the kind of high velocity, high repetitive tasks from that kind of lawyers do or legal clerks do, and then I'm going to automate them. And then is it now like, no, no. I can see how the organization works. So let me walk backwards from what is that kind of throughput of that vector of everybody working together? Let me think about the different atomic units of that. Let me automate those together. Was that the way kind of beginning and end, or from the beginning you guys already had the organizational thing mapped out?
- [14:44]
A
No, I think from the beginning it was very much the ideas came from the technology. And so I would think of the challenge we face of building the product is I think you still need to do this, right? Like, if you look at why the coding companies and the model companies build such great products, it's because they. They basically take the raw model and they put the least amount of things possible on top of it. Now I primarily use the models from the command line, which is essentially no product on top of the models, and you just interact with them. And so you have kind of what is the right shape of products, given the capability and the shape of the models? That's one constraint. And I would say that's kind of the bottoms up. And that's what you're seeing the labs and the coding companies do of how do I add stuff on top of these models that make it useful, especially for programming. And then there's the top down, which is, here is how a law firm works today, and here's all the things that you need for this to work in practice. And so when I think of product development, I kind of alternate between these two. Like, I use the models for coding all the time. And then we spend a bunch of time talking to customers, especially lawyers, and just saying, like, how do you work on this client matter? What do you do? What tools do you use? And then thinking about how to bring these closer and closer together, and they're still really far apart. One way I get a lot of product intuition is I try to do the work that our clients do with the code models. And so a lot of what I spend time doing is if I wanted to automate a diligence or I wanted to do a first pass of a diligence, how would I do that with the RAW models. And so for example, one thing I did a couple months ago was take all the documents you would expect if you were buying a company that would be in a data room, convert them to markdown, and then figure out all the skills and scaffolding you would need for an agent to read all those and generate a diligence memo. And then you conversion control it. You can put this all in GitHub and you can figure out, okay, from a capabilities perspective this actually works quite well. And so now you have a sense of okay, this is what the capabilities are. And now you need to do the translation where it's like, okay, this is usable for me because I'm a software engineer. This isn't usable for most lawyers. And so you need to figure out like they're not going to use Git for version control. Like they store all their data here. This is like the interfaces they need to access those documents in Word and Redline. And so I think to me, I kind of try to separate how would you get these things to work really well independent of how do you make them easy for a user? And that's kind of, we have like Harvey Labs, which is like a research team that solves these problems and just figure out purely how do you get the capabilities?
- [17:24]
B
So this is fascinating to me because in a way I, you know, one of the, one of the shifts I've led was the kind of transition from product and design to more of a full stack builder. But there's a belief that you have like system builders that are building the infrastructure, there's researchers who are building the capabilities of the future and then you have full stack builders who can just pick it up and figure out like what the customer wants really well, marry it with the capabilities and then just take it to market.
- [17:50]
A
Yeah, and I think this is where it gets to me, very complicated because as you think of what's going to happen with these agent systems, like a lot like the product we're building for law firms is increasingly going to be how do we help your entire law firm or your entire in house department run on agents.
- [18:06]
B
And it sounds like you see the next big step as like organizational agents working together. So you're no longer representing Gabe, you're presenting a team. You're presenting like there's multiple agents working together. And that's a different level of collaboration that is needed. Is this something you're tackling yourself or you're waiting for capabilities to show up to help you do that?
- [18:28]
A
The way I would think of this for Law firms, increasingly, the direction's going to go is if I'm a law firm, I'm just working on, say, a thousand different client matters. And what I really want is this infrastructure that for every client matter, as a partner, it lets me delegate this work either to an associate or to an agent. Those go off and do that work, and then they tag me in when they need my judgment. Right. And so what is all the, like, process management and infrastructure you need to enable that at scale? Well, and then we also, we already have a product called Shared Spaces where you can collaborate with your client. And so you actually have this workspace where that client matter is happening between the law firm and the client they're representing. And so to your point, the permissioning and how you ensure the auditability, the quality, the, you know, all of this that you need to do that work, I think this is some of the, like, very complicated, both product and technical problems that we're solving to enable this. But I think the capabilities are there.
- [19:28]
B
When you start to work agents to agents, is there a different layer of efficiency you can work because the agents just work with agents. So you can just talk in zeros and ones.
- [19:38]
A
I think this is maybe where legal is a bit different than programming. But if you think of what a contract is, at the end of the day, this is like an agreement between two parties, and I do need to understand what that agreement is. And so when you think of one of these client matters that these law firms is working on, it is either a litigation where at the end of the day, what I'm producing is motions that I'm going to present to the court, so those do need to be readable, or contracts where I'm buying a company and here's the purchase agreement, and I do need to align with the other party on that. And so I think one of the big problems that we are trying to solve, which I think does exist now in engineering, is the bottleneck is human review. Right. Like, we see this in our engineering. Org, where now everyone can write so much code that the bottleneck becomes how do our senior engineers ensure that all that code is correct? Just the same as, like, when you were running a team and you think of the best people that worked for you, it's like they were able to, like, distill their questions and what they did in a way that you were like, I don't need to go look at everything you did because I trust you. And it's like, we need to solve that. And if you think of what the best associates do for a partner. It's exactly that. I read the whole data room. Here's all the issues I flagged, here's the things I'm uncertain about. Here's my draft of the memo. And the partner reads that, checks a couple things, and they say, I'm confident enough to now go and give that to the client. And it's like, we need to solve that. And I don't think that problem goes away because at the end of the day, this client is paying this partner a lot of money to be able for that partner to say I'm willing to risk my reputation that this transaction is correct. And so I don't think there is in the short term a magic solve for that through AI. I think it's still the same problem.
- [21:25]
B
It sounds like it's less about can I get model to do legal work. It's more about how do I get them to be fully autonomous in a way that allows conflicts and resolution really well or matter subject folks accelerate it really well. Permissioning, defensibility, audibility, readability. I'm trying to imagine how your team looks like based on that.
- [21:45]
A
So we have kind of like an assistant team, a vault team, workflows, shared spaces, like a bunch of teams, like Embedded experience, which is building kind of Word and Outlook integrations, things like that and integrations team. I think a lot of it is mapping to like I would say the vertical teams map to the product surface areas. And then those teams are a mix of PMs, designers, lawyers, engineers, and then AI folks, if that product surface has an AI feature. And then the horizontal teams are also what you would expect of, you know, here is the team that builds a lot of the infrastructure, that abstracts away the different models. There's a team building the cloud agent infrastructure. I think right now we're kind of going through this transition where a lot of these teams made sense when the paradigm was using these trap models. And now that we're shifting this kind of from chat models to these agent type systems, I think there is a lot of these boundaries where one of the things we're thinking through is are these the right org boundaries? And then as increasingly you're growing, I think there's a question of, I think as we get bigger, do you want to map your product, your teams to product surfaces or closer to outcomes? And so I think there is like this really challenging org design problem. We have fdes, we have legal engineers which help customers deploy these workflows.
- [23:07]
B
Is this a model you see going forward? Like every customer Is again, they're all unique. So we're going to like there's adaptation of making Harvey work really well, 100% well for everybody. Otherwise it will be 60% well or 70% well. So it's kind of the last, I
- [23:25]
A
would say like our model for fd and maybe the name is a bit of a misnomer. It's not kind of the full Palantir model of we're going into there code base and building unique software for them. It's very much okay, we're working with a lot of large private equity clients. How do we kind of build them the product features that pull ahead our roadmap a bit and start merging this into the product. But I would say a big challenge we are going to help customers solve is I think this technology is very difficult to implement, especially at these large enterprises. But for example, with some of the large banks, we're starting to think about how do we automate large parts of their internal investigations. Right. And this is just an incredibly complex process that you do need to go connect this to systems, figure out how to coordinate the teams and the platform if you're working with a law firm. And so I think there is just a bunch of implementation work that you will need to do if you're in industries like we are. So I don't see that going away.
- [24:21]
B
I've read somewhere you can tell me if it's correct that you guys have a 92% monthly adoption rate, which is remarkable. Is that what you're looking for? How do you think of this at this point?
- [24:34]
A
I think we are a bit unique compared to maybe some of the other companies is I think there's actually very clear metrics that we want to optimize. So for a law firm, at the end of the day, the metric we want to help law firms optimize is profits per partner. My mental model for law firms is they are actually kind of a collection of business units. And so all of the different practice areas in a large law firm operate somewhat independently. Like if I'm doing IP litigation at a law firm, that business unit is separate than I'm doing fund formation for private equity clients. And so I think it's so complicated to move this number because what you actually need to figure out is how do I transform each of these practice areas? And each of these practice areas is different. Right. But you can start decomposing this. Right. So you say, okay, profits per partner is really the sum of all these practice areas and how profitable each partner is in each of Those practice areas. And then you can go look at each practice area and start thinking about, okay, what do my private equity clients care about for fund formation? I think people think of law firms as the vertical. But even within those, each of their practice areas maps to an enterprise vertical. So private equity has fund formation and ma, there's insurance, which is a separate. And so you need to go and solve all of those. And then for enterprise, you have the same thing, right? Like if you think of a massive enterprise, at the end of the day, they care about what is the quality of the legal work that my legal department is doing, how quickly are they doing it, and are they doing it cost efficiently. So I think a big problem that these vertical AI companies will solve is how do you start quantifying all of that? And you can't just do it, I don't think by, oh, everyone uses our product, right? Like, that needs to be built directly into the product. So you're measuring all of that and tracking all of that.
- [26:25]
B
So it sounds like in many ways like the true north is clear. But that's really hard to measure. So then you trickle it down to a lot of metrics which are very specific to what you're trying to do. When you think about the future, and you seem to be very thoughtful about the future and how you build today. If you go, let's get far into the future, five years for me, it feels very far right now, or even 10. Like, do you, do you see future law firms very different? Put aside the lawyer, like the law firms themselves do see, because the lower jobs obviously will change tremendously. But do you see law firms become very different?
- [27:03]
A
And one thing I feel pretty strongly about is in the next five, 10 years, the highest paid law firm partners will get paid significantly more than they get paid today. And I think the simple argument is just, you are going to give these partners an incredible amount of leverage and so they are going to be able to apply their super valuable judgment to more client matters. And I think the same thing is going to be true about software engineers. If you look at what's happening with software engineering salaries right now, they're going up, right? Like the highest paid software engineers and AI researchers are getting paid significantly more than they were 10 years ago and even five years ago because of that same kind of leverage argument. Like, you still need humans still in those client relationships, but their leverage is increasingly not just humans, but also agents. And then law firms need the infrastructure to be able to work in that way.
- [27:56]
B
If you could solve one product capability you want or AI capability by just snapping your finger. It doesn't exist yet, but you're like, if I had that man, I can accelerate my roadmap amazingly. What would you pick?
- [28:10]
A
Maybe not a capability, but if I knew the answer of what is all this infrastructure for agents going to look like? That, to me is one of the big hard questions we struggle with.
- [28:20]
B
What's a product you did not think would be successful actually surprised you?
- [28:23]
A
A product that I thought other people, like, maybe the bigger companies would do sooner. But I love is granola, where I think it's like a very simple product, but I use it to record all my internal meetings. And there's something that, you know, maybe these other companies have the capability whisper and all these things, but just packaging it super nicely. It's something that I use every day.
- [28:44]
B
What's your favorite non software, non digital product?
- [28:47]
A
I really like Fiori clothes. I think that's. Yeah, that's like a lot of my workout clothes. And I feel like just from a brand perspective, like, I think they've also inspired us a bit of, like, how we think about our brand.
- [28:59]
B
And then for someone who is an aspiring builder, they're, you know, at high school right now. What do you think they should learn?
- [29:06]
A
I would say what I did when I was in high school and college is I just spent a lot of time reading. And so I think I just had a ton of questions and I would just go on the Internet or read books and try to figure out what do I want from life, what do I want my career to look like, what makes me happy, who do I want to be? And kind of just figuring out all those questions. And I think now when I think about how hard those questions were to answer and how good the models are getting at synthesizing all this, I think people have this amazing ability to figure out, what do you want to do? And I think the mistake that maybe not mistake, but the way that I didn't think about these things is I think sometimes when people think about their career or their life, they say, I want to help people, so I want to be a doctor. And then they kind of latch onto that thing and then they follow this career path and it ends up not being the thing that they want. And I think I always tried to think of these things more from first principles where I think it's right to start of I want to help people or this is the thing that I value. But then try to think about, these are the things that I enjoy doing. These are the things that I want to work on. Like, this is the things that I found myself drawn to spending time on. And I think the big challenge is, I think a lot of the really exciting jobs, if you're thinking about that now, don't exist right now. And so that to me is the big opportunity. And then I would say do that some of the time and then spend time with people.
- [30:34]
B
Love it. Gabe, this has been. I've learned so much from you, man. This just conversation has been super insightful. Thank you so much for sharing.
- [30:41]
A
Thanks so much for having me.
- [30:43]
B
There are many great takeaways from this conversation, but here are a couple which I think are quite unique to how Gabe and Harvey are building. First, they start with the model and then they design the product. True to his applied AI background, Gabe starts with thinking how to solve the need through the model. First, before he adds any UI features, he's trying to push the model as far as it can go to see what actually works. And then, without limitation, he builds the product around it. Most companies do the opposite. They take the model as is, and then they build workflows and interfaces on top of it. They're essentially model takers. Harvey is much closer to a model shaper. They start with the foundational model and then they expand what the model itself can do. And if they get that right, something really interesting can happen because the product actually becomes simpler. There's less ui, less friction, and there's more intelligence at the core of it. The second takeaway is about what it actually means to win in AI products for legal or for other areas as well. It's never been easier to generate answers or to take actions. That part is getting commoditized very quickly. In AI, what's much harder is delivering real outcomes with AI and standing behind them. In legal, the job isn't to generate documents. It's taking responsibility for those filings, for those contracts, for the correctness of them. Any AI tool can draft a contract, but what matters is empowering a partner to say, this is right and I stand behind it. That's the difference. This is where the real value is. The last takeaway is about where the real unlock is. As AI accelerates execution, the bottleneck is no longer building and shipping. It's deciding what to build and how it all fits together. When everyone can generate ideas and ship quickly, the constraint becomes coordination. So without clarity on priorities and alignment, more output can lead to more noise, not more progress. There's a lot of focus right now on individual productivity, but the real unlock is when the teams actually come together. Not just teams of humans, but teams of humans and agents working side by side. The products that actually start with that in mind will be much better positioned to win. In B2B. I'm Tomer Coyne and this is building one. Keep building.
- [33:25]
A
You've been watching Building One. Our show is hosted by Tomer Cohen. Building One is produced and edited by Mason Cohn and the team at Coastal Production Works. This episode was mixed by Tim Boland at LinkedIn. Our team includes Rachel Karp, Sarah Storm, Dave Pond and Alicia Mann, with support from Alex Kuznetsova and Mujib Merdad. Until next time, keep building.