Kyle Lawrence and Moish Peltz interview Killer Whale Strategies Founder Zach Abramowitz about real-world AI adoption in law firms, the risks of “tool obsession,” and why collaborative, innovative strategies are winning in legal tech. Learn how business models, attorney roles, and legal outcomes are changing because of AI.

Subscribe to Block & Order to learn more about the latest news on the legal side of digital assets!

Follow B+O on Our Socials:

Twitter: @BAndOShow
LinkedIn: Block & Order
Instagram: @blockandorder
TikTok: @blockandorder

Chapters:

00:00 – Welcome to Block & Order
00:11 – The Orca Movie & Naming Killer Whale Strategies
04:44 – Tech & Lawyers
08:20 – Introducing Zach Abramowitz
09:26 – What Law Firms Get Right (and Wrong) About AI
14:36 – Tool Obsession & Becoming “AI-Pilled”
17:04 – Harvey vs. Lagora: Comparing Legal AI Startups
22:35 – Anthropic, Foundation Models, and Legal Tech’s Future
30:47 – Legal Business Models: Pricing in the AI Era
44:26 – Predictions for the Future of Law & AI & Conclusion

Watch or listen to the podcast here:

   

Transcript:

**This transcript has been prepared automatically by AI and may contain inaccuracies**

Kyle Lawrence [00:00:11]:
So as we prepared to have you, Zach Abramowitz, guest on the show, I was gearing up to try to figure out how do I Segue Orca, the killer whale into the conversation. For those of you who don’t know, the famed 1977 classic starring Richard Harris and Charlotte Rampling is one of my favorite movies as a kid. Not because I’m fascinated by killer whales. And the fact that the killer whale in that movie, the titular whale, knows, like, thermodynamics and human biology and destroys the town. It’s amazing. And we get on the video, and I see you have the posters right behind you, and you completely took the wind out of my sail. So is that movie the impetus for the name of your company?

Zach Abramowitz [00:00:55]:
For the name of the company. Killer Whale Strategies. So there are. It was just so funny. Someone was asking me today about all my orcas in the background. So the name Killer Whale Strategies is based on the idea that if you ask most people what the spirit animal of the lawyer is, and they will tell you, it’s the shark, right? The thing about sharks is they hunt alone. And since the dawn of time, they’ve done very little iteration on their hunting method. Sharks are hunting today the same way they were in prehistoric times.

Zach Abramowitz [00:01:31]:
They hunt alone. And by comparison, the killer whale, because we think lawyers ought to use Killer Whale Strategies. The killer whale is unique in the animal kingdom that it constantly iterates on its method of hunting, even when it doesn’t necessarily need to. So it’s not like, you know, you know, the polar bear with, you see like the. The pictures of the polar bears, like, swimming now because of global warming. And what. This is not like a situation where it’s like, well, there’s been some, you know, problem in the ecosystem. Will we evolve and survive or will we not? The killer whale, even when it’s got very good hunting methods, is constantly looking to do better.

Zach Abramowitz [00:02:20]:
It takes big risks. You see, you know, this picture right here is a picture of a special pot of orcas off the coast of Patagonia in South America that knows how to hide inside of waves and come up right on the beach and grab seals, right? They. This method has been perfected within the last century. This is not something that is like something that they’ve learned, you know, thousands or even millions of years ago, where we’re talking about recent there and we’ve. We’ve seen it like in our own lifetime. There’s. Now there’s one that knows how to, like, go sneak inside, like the kelp as well, that’s been documented on BBC. They’ve started hunting great white sharks off the coast of South Africa.

Zach Abramowitz [00:03:07]:
So I think this is a great analogy for the future of lawyers. Number one, killer whales tend to, you know, work smart, not hard. They tend to be much more collaborative. They hunt as a pack, not alone. But again, I think their willingness to go outside and learn these new methods of hunting, you know, the killer whale only hunts. And I’m sure most lawyers were dying to get on this podcast and hear tons of, like, stats on killer whales. But the killer whale, I’m ready to go, dude.

Zach Abramowitz [00:03:35]:
Around 5%. Around 5% of its time hunting. 95% of its time is spent playing and interacting with, you know, with. With others in. In the podcast. So that was the inspiration for the name. Now, personally, when I was younger, I happened to watch this, this 1970s B movie, Orca, the killer whale that came, you know, in the wake of Jaws. And I think that whoever came up with the idea was like, well, Jaws did so well, let’s just kind of do the same thing with a killer whale.

Zach Abramowitz [00:04:10]:
And as you mentioned, the movie was so preposterous. It was like Moby Dick meets Die Hard, right? So you have, like, this, like, killer whale whose mate has been killed by, like, you know, weird old Captain Ahab. And he vows revenge on this captain. He’s able to, like you said, he knows thermodynamics, able to set off, like, this chain explosion that destroys the sleepy little dock town where the captain is staying when he’s not in. He’s not at safety. And the movie was totally ridiculous. Now when I watched it when I was 4 years old on a Saturday night because my sister and I were just up watching movies and our parents were nowhere to be found, we.

Zach Abramowitz [00:04:56]:
All of those, like, nuances were lost on me and I developed this horrible fear of killer whales. But by the way, it’s so funny. People of our age have all seen this movie because it was, you know, such a bust that they were basically giving it away to, like, you know, free TV to show on Saturday nights to unsuspecting school children. So we watched this movie. I became terrified of killer wells to the point where, like, my mother thought it’d be a good idea to send me to watch Free Willy. And I’m the only person who spent, like, Free Willy, like, sweating bullets and screaming, like, any time Danny fell in the pool, I’m like, oh, my God, he’s going to get blown up by that orca, the killer whale. So like it was. What I’ve learned over time is replacing fear with fascination and enthusiasm is just the, that’s the winning method.

Zach Abramowitz [00:05:47]:
Right. And I think the same thing is true when it comes to lawyers and technology. Like, I’m afraid of new technology too. I’m afraid of having to change. I’m not like someone who necessarily looks forward to new things or having to make huge changes in, in my life. But I think if you can replace fear with fascination and excitement, you can, you can really reframe how you view the whole thing. So I think it’s both, both my, my killer whale posters are deeply symbolic and I, I. And hopefully good illustrative examples for other attorneys for sure.

Kyle Lawrence [00:06:28]:
Well, I certainly feel seen and I share your fascination and I completely agree with you.

Zach Abramowitz [00:06:32]:
Toad Eric was also in that movie, by the way. It had some famous people.

Moish Peltz [00:06:38]:
So my favorite, I’ve never seen the, the John Wick of orca films.

Kyle Lawrence [00:06:42]:
So I mean it really is.

Zach Abramowitz [00:06:44]:
I ordered it a few years ago back when like Netflix was still doing DVDs. I was like, listen, I gotta like come face to face with my fears only to find out like half the footage of the killer whale was like filmed at Sea World.

Kyle Lawrence [00:06:55]:
Yes.

Zach Abramowitz [00:06:55]:
You can very clearly see these are not ocean waters.

Kyle Lawrence [00:06:59]:
You can. But, but to your point. So in that movie The Orca kills a great white shark brutally. And then the next year Jaws 2 came out and the great white shark in that movie kills a killer whale. So they had the studio tit for tat thing going on.

Zach Abramowitz [00:07:15]:
Very, very interesting now. You know, like the, in recent years they’ve, they’ve shown that this is like a, like a non battle. Right. Like the.

Kyle Lawrence [00:07:22]:
Yes.

Zach Abramowitz [00:07:22]:
The killer, the killer whale kills the great white shark in something like two seconds.

Kyle Lawrence [00:07:26]:
Do you know how they do it? Moish? I promise we’ll get to like actual stuff. So, so the way the killer whales do it, it’s amazing. It knows to turn the shark upside down, puts it in what’s called a state of tonic. And when the shark is upside down, it can’t move. And they then proceed to eat its. Or like for fun. They’re not even hungry. It’s mind blowing.

Zach Abramowitz [00:07:44]:
Yeah, I feel like this is an animal lawyers can get behind.

Kyle Lawrence [00:07:47]:
Right, I’m with you man.

Moish Peltz [00:07:49]:
I’m with you now. I didn’t understand the reference and now I’m sold. Like, great, great call out.

Kyle Lawrence [00:07:55]:
You can go watch the movie later with your kids. Surely they won’t be scarred for life by watching it. That was by far the best introduction we’ve ever had here on Block and Order. But duty calls. I do have to formally introduce our guest at some point in our one hour runtime. So our guest, I believe, needs no introduction. Zach Abramowitz is one of the most recognized voices at the intersection of law and technology. He’s the founder of Killer Whale Strategies.

Kyle Lawrence [00:08:20]:
If you did not glean that from our opening, where he advises law firms and legal technology companies on go to market strategy growth and navigating the rapidly evolving AI landscape. Before launching Killer Whale Strategies, Zach built a reputation as a sharp commentator on legal innovation through writing, media appearances and his own podcast work. He’s been named to the Fast Case 5050 list, honoring legal innovators and his insights on how AI is reshaping the legal profession and have made him a must follow for anyone paying attention to where the industry is headed. Zach, welcome to the show, my friend.

Zach Abramowitz [00:08:52]:
Thanks, guys. Thanks to, Glad to be here.

Kyle Lawrence [00:08:56]:
It’s good to have you on. So as much as I want to talk about killer whales for the rest of the show, you write a lot about lawyers and utilizing AI. What do we get right? What do we get wrong as a firm? As our firm has really embraced AI and is approaching new technology with open arms and with the fascination that you mentioned in the lead in what do lawyers get right and where do they, what do they get wrong in implementing AI into their daily practice?

Zach Abramowitz [00:09:27]:
So, first of all, I think, you know, the, we’re still at the early phases of AI, and I think that for a lot of people, if you’ve been in it for, you know, three years, maybe initially as a spectator, maybe more recently as a user, you might feel like, okay, we’re at, you know, the, the beginning of the end. And I think we’re more like at the end of the beginning. Right? So I, I, it’s, it’s hard already necessarily to talk about what lawyers are doing right and wrong with AI because we’re still experimenting. I think the, the thing that would have been wrong is if over the past three years, you stuck your head in the sand. You tried to pretend that this trend would go away, that we were in a bubble, that we’re, oh, it’. This is, this is just a phase. This is like the metaverse or this is like, you know, web 3. To pretend like this wasn’t the most significant development of our lifetimes, I think would have been the absolute wrong path.

Zach Abramowitz [00:10:35]:
Now, the good news is I think that, you know, when it comes to the data, you see that lawyers actually have been adopting AI. And this has not been one where they’ve really tried to sit on the sidelines. You know, whether you saw that. I think already in 2025, there were more lawyers, according to Law360. You know, this is from the beginning of 2025, there were more lawyers that were using AI in their daily work than were not. And the numbers had previously been, you know, in 2024, only around 30% of lawyers were. Were using AI at work. So you see already that there’s been a shift and uptick in adoption.

Zach Abramowitz [00:11:17]:
Last year there was a study done by Wharton at the end of the year that showed that the greatest, like, change in adoption usage in large enterprises had been the legal function, you know, with 24 percentage points change. So it is. I think we have to. First, before we talk about what lawyers have, you know, done wrong when it comes to AI, I think it’s important to note that, like, lawyers have done a lot, right? I mean, if you think about, like, Harvey and Legora and these valuations that they’re getting in the billions of dollars and the fact they’re generating hundreds of millions in revenue, that doesn’t happen without, like, law firms and lawyers actually subscribing to these solutions. And lawyers have tried to sit out a number of technological advances and I think successfully in some cases. I think lawyers, for the most part, like, sat out CRM. And did it really, like, affect them? No.

Zach Abramowitz [00:12:18]:
Right. So I think there’s plenty of. By the way, I think even in the last, you know, 10 years, if you could have, you could have still done your deals not using, you know, Kira Systems or eBrevia and still done fine, maybe better. And I think the. There’s. There’s definitely been a shift, and I think we have to recognize the adoption. Now, where I think lawyers have gotten things wrong is in a few areas. First of all, the obsession with tools, right? I tell lawyers, like, you know, your innovation strategy does not boil down to, like, Harvey or Legora.

Zach Abramowitz [00:12:57]:
And before that, by the way, it was Harvey or co counsel or it was co counselor. It’s been a different, like, tool obsession constantly. And again, if you’re, if you’re in, like, you know, firms outside of big law, it’ll be different tools that maybe you’re looking at. But it ultimately comes down to, like, which tool should I use? And my feeling is, like, pick one and go with it, right? The worst thing you can do is get stuck in analysis paralysis about what product that you’re going to use, because the most important thing, and I think this is kind of leads to the second thing, that lawyers may not have been done right enough, but I think we’re starting to get there. The most important thing is to kind of get yourself AI pilled. And that can be done with any tool, right? A lot of times people will say to me, you know, oh, I’ve seen Harvey, and it’s not really that impressive. All it is Chat GPT. And I’m like, you mean Chat GPT, like, the single most useful product you’ve ever used in your life? Like, that’s all that Harvey is.

Moish Peltz [00:13:54]:
It’s just a rapper. It’s like, oh, it’s just a rapper.

Zach Abramowitz [00:13:57]:
Are you crazy? Crazy. Like, if. If all it was was making Chat GPT like, accessible and available to big law firms with literally tens of thousands of clients and active matters going on at once, that would be extremely impressive. So you got to just use a product so that you start, like, realizing, okay, here’s what it can do, here’s what it does, doesn’t do as well. Here’s how I interface with it the best. And I think that using the AI to kind of AI pill yourself is. Is really important. I think the other thing, and this kind of bleeds in, you know.

Zach Abramowitz [00:14:36]:
The third thing I think lawyers have done wrong is just the obsession with, what’s the use case? What’s the use case? And when I hear someone asking, what’s the use case? It’s like they’re saying, we have this, like, one single problem. We’d like to beam the AI at that problem. And then for the rest of the day, we’d like to do things exactly how we were doing them before. No changes at all. And that’s exactly the wrong thing to be doing. What you should be doing is learning how AI sort of functions as your second brain. Sometimes you need your second brain, sometimes you don’t, right? Like, tell people all the time, like, we drive with Waze, right? Or with Google Maps or any GPS system, right? You use it when you need it. I don’t use it when I drive to, like, the local supermarket.

Zach Abramowitz [00:15:29]:
I know how to get there, right? So I don’t. Oh, I don’t use it for every single thing I do, but it’s available. It’s always there for me if I need it. And as a result, I don’t. Like, when I drive, I have, like, a different level of confidence about my ability to get anywhere I need to get. And I think that’s like, a similar thing with AI. When you start thinking about it much more as a second brain, as a helper. I like to point out a lot of times that AI got much more popular in the wake of the COVID virus.

Zach Abramowitz [00:15:57]:
Right. That was like, Covid. When, when Covid kind of finally ended, we’re like, okay, everybody doesn’t have to wear masks. There’s really no luck. Okay, it’s really over. All of a sudden, AI appeared. And I think the timing is not coincidental, meaning, not that they timed it for that, but part, part of the reason it grew with such fervor was because people had been working alone and the ability to discuss something with, you know, John or Jane down the hall now was not as simple. And very often you were in the place of having to work through something on your own.

Zach Abramowitz [00:16:33]:
So all of a sudden you had this like, you call it personal assistant, I call it thought partner, you know, whatever it is. All of a sudden you had someone that you could work with, that you could talk to at a moment’s notice without scheduling a zoom. So I think getting into that, that mindset about AI, you know, kind of AI pilling yourself, I think is also really important and not just to think about like, okay, we use AI, like for our data rooms.

Moish Peltz [00:17:04]:
Well, you recently had on your podcast, you know, speaking about the paralysis analysis and tool selection, you had Winston from Harvey and Max from Legora on your podcast. And I’m sure in your day you speak to other founders of legal tech products and try them out. So I’m curious in that, what do you see differently about these two companies which are now multi billion dollar valuations and have two young entrepreneurial founders, like, what, what do you see differently about the founders and these products in the marketplace? Are they really just, you know, we have Coke and Pepsi and you can pick one and drink it and it’s good and it’s, you know, we’ll go from there, or like, you know, are they approaching this problem from different? I’m just curious what you see, you know, having having a bit more insight into this industry.

Zach Abramowitz [00:17:52]:
Yeah. And I’ve changed my mind a bit about this. So first of all, I, I, I was very intentional in separating Harvey and Legora by at least one episode. I actually had wanted to separate them by more, but just based on the timing of when we had to record, I ended up separating them just by, just by one single episode. But that was symbolic because you could definitely take away the like the Coke vs. Pepsi analogy. And I think that that really is at the moment, if you’re talking about the Amlod 200. If you’re talking about major Fortune 500 departments, I think if you’re even talking about, you know, law firms with, with 50 attorneys or more, you probably, you know, like Coke and Pepsi, Harvey and Legora probably are the two biggest names now. I think there are others.

Zach Abramowitz [00:18:40]:
I think you’re starting to see a number of Dr. Pepper snapples move into the space as well. But I’ll tell you something, and this is the, you know, the quote I, I mentioned most often from my podcast comes from a podcast I did very early on in sort of in the life of this podcast. My podcast Legally Disrupted with John Levy from Y Combinator. Right. So John Levy and Carolynn Levy are two of the most humble people in the world. They have handpicked most of the legal tech companies that have come out of Y Combinator. They were previously both attorneys themselves.

Zach Abramowitz [00:19:19]:
Carolynn’s the head of legal. Carolynn actually drafted the safe agreement, created that very concept for startups like you. Talking hugely influential people, but extremely humble. And I asked John on the podcast, I said, hey, listen, you know, I’ve noticed Y Combinator has always been investing in legal, but you’re investing even more these days. Is that because you’re particularly bullish on AI as it like, you know, applies to the legal sector? And they said, he said, no. He said, actually we don’t really think about verticals at all. We are like the NBA draft. We simply try to pick the best talent on the board.

Zach Abramowitz [00:20:01]:
That’s it. And he said, however, it is the case that more top talent today is trying to build and legal than ever. So that like our, our entrepreneurs are paying attention to certain, to certain sectors and going after them. So, you know, I think take someone like, you know, like Max, like Max, unlike Winston, never practiced law at all. I don’t think in, you know, in a previous decade, Max probably starts a company, but probably not one in the legal AI space, right? Because it just might not have been as attractive. Whereas today he’ll tell you, like, this is his life’s mission, this is what he plans to spend the rest of his professional career doing. That is unique and that is very new. And I’ve been investing in companies in the legal space since 2019.

Zach Abramowitz [00:20:54]:
And today it’s like the, the simple, like simply the number of top talent teams who know how to execute at the highest levels is just unmatched. And that you can follow this like down the line at like all of whether it’s like Harvey Legora Trial Kit New code, solve intelligence. You go down, go down. And, you know, with all these companies, part of the reason they’re doing so well is not just because the lawyers are getting on board, but it’s because they’re much more effective at selling, marketing, building products. And then, of course, the technology you have today is simply like nothing it has before as, as it pertains to Harvey and Legora in particular. Where I think I’ve really changed my mind is I, I kind of saw them initially as. Oh, they’re just, they’re going to be the same kind of company, and some are going to choose Harvey and some are going to choose Legora. And it’ll be a relationships business.

Zach Abramowitz [00:21:53]:
It’ll be maybe who’s got the better brand. And I think that we’re actually going to see something different. I think that you’re going to see these two companies take very different paths. There’s, there’s a number of, of features already where they’ve looks like they’re doing the same thing, but not necessarily so, you know, like, there’s no secrets, like, everyone, like, if you’re, if you’re on Team Legora, you know what the most recent version of Harvey has and vice versa. And so it could be a game of, like, copy and paste, but I actually think you’re already seeing, like, the two founders with slightly different visions that I think will become much more clear over time.

Moish Peltz [00:22:35]:
Yeah. And then how does that compare? I mean, we had Zach Shapiro on the show, and I know, you know, we’ve all through X or different podcasts, talk, talk to each other. His, his view is very strong that all the value here eventually just accrues to Anthropic and they’re going to eat the world, so to speak. I don’t know. It’s a more nuanced view than that. But how do you compare, you know, what’s happening at the foundational layer to what Harvey and Legora are doing? And, you know, I think obviously from the AMLA 200 perspective, you know, there’s certain needs. It’s having an account executive. It’s things that Anthropic may not be prepared to provide you at this time.

Moish Peltz [00:23:14]:
So there’s value in that relationship aspect as you describe, but over time, that might erode and Anthropic might be able to scale, you know, their account reps. Like, how do you think about that? Not just where it is today, but, you know, three to five years down the road, or, you know, Coke has been around for a hundred Years. You know, these, these are very, you know, these companies have the same shelf life.

Zach Abramowitz [00:23:36]:
Perhaps first of all, it’s possible, right? Meaning that, that, that view that essentially, ultimately the frontier models are just going to consume everyone and eat everything, right. I’ve sort of described this as the Hungry Mungry problem, right? Hungry Mungry is a poem by Shel Silverstein and comes to dinner, eats everything and eventually, you know, eats himself. Right? And to a certain extent, like it always feels like AI just eating everything. And then I go outside and like everyone still has a job and everyone’s still super busy. I go to firms these days. I love to ask, I’m like, I’m like, hey, by the way, how many of you are not busy since I came out? And like, like no hands go up, right? Who? Everyone’s busier, right?

Zach Abramowitz [00:24:24]:
So we’re busier. You know, I think the, the sort of our, our sort of lived experience as well as the data, you know, suggests that maybe AI isn’t going to eat everything. But I, I think that that view is like, let’s say that is the prima facie view, right? Which is. Yeah, I think like at first blush, will the models own everything? Yeah, like, we saw this, we’ve seen this with, with social media. Remember Zynga? Remember all the companies that tried to build on Facebook. Do you remember Clubhouse or Twitter? Like Clubhouse is just consumed by Twitter spaces, right? Meaning like I get, by the way, my first company reply all was built is tried to be like, as kind of an early version of Clubhouse. And what we saw is it’s just, it’s so. It was so hard to compete with anything that was even like a imagined potential feature of a, of a social media platform that it’s almost like not worth trying.

Zach Abramowitz [00:25:25]:
And by the way, there’s plenty of other examples of this even in software. The. Plenty of companies that, you know, existed and were around for a while until Salesforce just made them totally irrelevant. So I def. And sometimes companies would partner with Salesforce so build on the Salesforce ecosystem and like ultimately that it could definitely go in that direction. But I think things rarely work out neatly. Right. And I think that there’s like, where I see that view and I understand it on paper, I definitely don’t know if it will work out that neatly.

Zach Abramowitz [00:26:09]:
I asked Max Strong recently on the podcast whether he thought that Legora is a trillion dollar company and he says yes, he thinks that’s what it can be. And the theory is that Software today is just simply much more valuable than software was 10 years ago. And I think we’ve seen that with AI. I mean, ask yourself, like, I cannot think of another tool. I mean my, my only kind of useful comparison is GPS in Waze because it just fundamentally changed the way I drive. Whether that’s me driving or whether that’s my, you know, trust to take, you know, an Uber or a Lyft and just be confident no matter what, even if the person has no indication they’re a taxi driver, that they know how to get from one spot to the other. That’s only possible with something like GPS. So I think we’ve seen it before.

Zach Abramowitz [00:27:05]:
But I think the other piece of this is there’s not just one frontier model. And this is a point that Maor Shlomo, the founder of Base44, which is a company that competes with Lovable and Replit AI coding. And really one of the, one of the very interesting stories of AI over the last two years was base 44 and its acquisition by Wix. Even though like it’s not, it’s. It’s Wix is probably going to become Base44, not Base44 kind of getting merged into Wix. And I think what, so what he says about this is if there’s only one model that pulls far ahead of the others

Zach Abramowitz [00:27:55]:
then we’re, then all of the application layer is in trouble. But as long as the models have to compete with one another, then there will always be an ability to build on top of the models and build further application layer. Because anthropic sometimes, I’ll say that Harvey, they’ve made this point, they’re marketing that their bigger competitors have to be worried about are the frontier models. That’s true for the frontier models also. And especially the frontier models really need to be concerned about what the other frontier models doing. Everything today has been built off the paradigm of the transformer model. And the transformer model is, is incredible. But what if someone comes out with something new? What if someone comes out with a, with a, with a new model that does something where we go oh wow.

Zach Abramowitz [00:28:50]:
And we start looking at transform models today the way we look at recurrent neural networks which, you know, were kind of useless, where the transfer model made it like extremely use, you know, made, made AI much more useful. Maybe we see a different kind of model ultimately that is either much more efficient in terms of compute or takes kind of a different tactic. You know, there’s, there’s going to be a lot of crazy stuff coming out of AI research labs. So I think that the, that’s what the AI models have to have to focus on more.

Zach Abramowitz [00:29:28]:
Sure will they throw out a couple of like, press releases about legal so that more lawyers use them? Of course. Do I think that like, there’s like they’re sitting aside in like the, you know, the, the private chambers of anthropic going like, what do we do about that, Harvey? No, I don’t, I don’t think it’s. I don’t, I don’t think it’s. I don’t think it’s really, I don’t think it’s really, really on their radar just yet.

Moish Peltz [00:29:55]:
Well, I think, I think that’s actually again, one of the best bull cases for Harvey and Legora. And the way we think about her is it takes a lot of time to train employees on one of these things, let alone, all right, let’s use Claude this week and now let’s use OpenAI. And now let’s use Gemini. And I’m sure it’s going to change right back and forth. So the fact that they aggregate and have a consistent workspace and you can drop in the model that is performing best this week is a huge time saver and value add for our firm, you know, but that, but then the switching costs are definitely tremendous. And so I, I think that’s a great example of that. Now one thing that I, I think is, is I’m curious your thoughts on are the changes in business models. Both, both for, you know, Harvey Legora, which I’m hearing through the grapevine, are increasingly.

Moish Peltz [00:30:47]:
I think Legora just announced last week that their new future, I think the agentic future is moving to a consumption pricing model.

Zach Abramowitz [00:30:56]:
Yep.

Moish Peltz [00:30:57]:
And as I understand Harvey is also must be contemplating these kinds of things. And then I think law firms also are now having to think about, well, we bill by the hour, but if we can do, you know, 40 hours of work in 5 hours, we’re not going to 5x our hourly rate. We have to do value pricing or subscription pricing or some sort of.

Zach Abramowitz [00:31:19]:
Although they might 5x their hourly rate. Right. Meaning that that actually wouldn’t be crazy. Meaning I do think there’s. There was an essay written about this called the, the High Density Hour by an attorney at Wilson Sonsini where he essentially argues that that work that is like highly sophisticated and unpredictable and high stakes will probably continue to operate on the billable hour and that the human judgment will become more valuable in like, in density, essentially. So I don’t think it’s I don’t think we necessarily move away from billable hour. I actually think that in a world where you can create the deliverables very easily, that the level of human judgment becomes more important. So very often these days I’m actually billing on billable hours now, not billable hours where I say I’ll figure out how long it’s going to take me and then I’m going to, you know, I’m going to count.

Zach Abramowitz [00:32:25]:
But what I have done is said, listen, I am going to spend 10 hours with your materials. If I wanted to create the deliverables that you’re asking for, I could create them in five, six minutes. Wouldn’t be difficult. But anyone can do that. So what I’m going to do is spend a significant portion of my time all attention. Like I will give it a 10 hour working session where I’m entirely focused on what you’re doing. That’ll be a mix of research, that’ll be a mix of me thinking, that’ll be a mix of me writing, there’ll be a mix of putting slides together, whatever it is. But I think that there is, there is a significant added value.

Zach Abramowitz [00:33:06]:
So I don’t think that that’s outside the realm of possibility. At the same time, I do think that you’re seeing a lot of like, potential changes in business models. I think the sort of the most popular one you mentioned, Zach, before, is this quote unquote, AI first firm. Now I think that the term AI first or AI native was very useful originally. I think it’s losing some of its meaning. Like, what is an AI first firm? Right. So I sort of defined it as a firm that, like, how would you build a firm today if you, one, knew that AI existed? Right. And two, like, how would you build it slightly differently or maybe very differently than you would have built the exact same practice five years ago.

Zach Abramowitz [00:33:56]:
Right. So I think that that’s sort of how I define it. I think you can go in a bunch of different directions. And in some instances I’m really bullish on the AI first firm model. And in some instances I think I’m not really sure, like, and especially I don’t always understand why they need huge amounts of funding. Right? So like if, if the whole point is AI exists, AI exists. And now because AI exists, you build a firm much differently than you would have. Well then, like, why do you need to be going and developing your own AI and raising money to do that? So I know there’s some firms that are raising money in order to Hire attorneys.

Zach Abramowitz [00:34:38]:
But my point is then shouldn’t the attorneys see the opportunity of the AI first firm and be willing to get in on that and want equity in that and want the upside? Like why does it then take, why does it take, you know, venture capital just to hire people? Doesn’t that sort of speak to like lack of confidence in the model if they won’t like go and just sort of do it themselves? Now I do think there are exceptions to that. I think there absolutely are exceptions. I think that the more you get to like niche workflows, the more you get to something where you can say, listen, this firm is literally only going to do one thing, but that one thing we’re going to do better than anyone else. And there’s a huge portion of that that is tech and process driven and it’s going to require a serious amount of development. Then I do think that there, that there’s, that it makes some sense. I think maybe if you’ve got like a PE roll up strategy with firms where you want to go out and buy other practices and bring them in, I think it could make sense to raise capital. But I don’t quite understand why all of these firms are raising money. Especially when it comes to things like, oh, we’re a firm that now does like your seed round documents.

Zach Abramowitz [00:36:02]:
It’s like, did we need like to call this AI first? Seed Round documents have been like template for a while and have been something that like serious attorneys don’t really want to deal with because it’s like how much, how many issues could there really be at the seed? Usually you’ll have like some junior attorney do the work. So I, I think there’s, there’s work that like was already on its way to being like commoditized or flat feed and was already sort of leaving the big firms. I mean I, they’ve had like, some of the big firms have like had places where you could just go and like fill out a couple of forms and it would produce like your, your series A documents or something.

Kyle Lawrence [00:36:41]:
Start up in a box.

Zach Abramowitz [00:36:41]:
Oh yeah, yeah. Start up in a box. Exactly. Some, some, something like that. So to pretend like, oh wow, this is, this is so new. Like I, so I just interviewed Dan Mishin, who’s the founder of Manifest OS, which is like specifically for corporate immigration. Now there are other firms that specialize in corporate immigration, but I’m just saying like that’s one where I’m like, okay, that I hear a little bit more because you’re talking about a very like specific Niche, something that like Harvey and Legora or anyone is not really going to be working on per se, and there’s not so many customers to sell it to. So in that case, it does make sense to go and start like a different kind of firm.

Zach Abramowitz [00:37:24]:
And I do think that there’s a huge opportunity for people like Zach and others, and I know people are doing this right now, to go and start firms today and say, hey, listen, I might not have started my own practice once upon a time, but today that actually feels more doable than ever. So in that respect, it’s, it’s AI first. And you’re probably not necessarily going to hire a huge number of associates that you’re just levering up matters with or you’ll do. You’ll do things a little bit differently.

Moish Peltz [00:37:55]:
But it sounds like you think the big firms can, instead of this being a revolution for them just on the margins, they can maybe adapt their leverage models a little bit and they’re going to be fine. And, and like now.

Zach Abramowitz [00:38:08]:
Totally. And I. But I think so, first of all, yes. Second of all, I think that there are changes that can be made. So for instance, one thing I’m sort of bullish on is junior attorneys being able to take on significantly more responsibility much earlier in their career. If you’re a high agency person, I mean, you’ve got AI. like I remember, like, you know, if it used to be the case we started a big firm, you were. If you just had to kind of know that there are certain partners that you should gravitate to because they’re better and they’ll mentor you and they’ll teach you and hopefully that partner will stay at the firm long enough to help you make partner.

Zach Abramowitz [00:38:52]:
Right. Luck of the draw. If you didn’t get that partner as a mentor or if that’s not who you were assigned to, or you just didn’t know that that’s what you’re supposed to do. Right. Which a lot, I think a lot of attorneys don’t really know how they’re supposed to succeed. Succeed in firms when they, you know, when they go there. So that would have been challenging today. It’s been democratized because today, like, if there’s something you don’t know how to do as an attorney, you should be able to look it up and, you know, might you be doing it exactly the way that the.

Zach Abramowitz [00:39:20]:
That assigning partner wanted you to do it? I don’t know. But like, you, you can, you can get smarter much more quickly. I think one thing that you might see Firms start to do. I know everyone talks about like, oh, we’re just going to stop hiring juniors. I think it might be that you stop keeping senior associates around for as long as you do because you might be able to get much more work out of the juniors. You might be able, you might have a meaning. It could be, and this is where I say things tend to not necessarily work out neatly and nicely at all planned. It could be that the move for general counsels is going to be, instead of saying, hey, we’re, we’re not, we don’t want to see any bills from your junior lawyers.

Zach Abramowitz [00:40:05]:
They might say, hey, listen, please stick. Please take juniors on this for more. We might see juniors making. We might see lawyers making partner after four and five years simply because of their capabilities with AI.

Kyle Lawrence [00:40:18]:
Yeah, it’s really interesting. I mean, usually our clients, it’s the other way around. They want to see the juniors on the bill. But a lot of concepts, you know, coming together and crystallizing things. We’ve talked about the, you know, what is the billable hour? What are our junior associates doing? All the information flowing to big law with so much movement in the space and so many new technologies and new programs being offered all the time. I mean, you’re much more plugged into it than. Than I am and, you know, Moish is too more than I am. But how do you sift through the flotsam and jetsam of all of this? How do you, how do you advise firms as to what is right and what is wrong? I mean, at the very beginning of the episode, you talked about saliently, I think more is lost from indecision than from the wrong decision.

Kyle Lawrence [00:41:01]:
So get off the pot time. But where do you start if you’re one of these, you know, young firms that’s trying to get in and do things differently? Where do you even begin with something like that? Do you just not have 0.1 billable hours? I mean, like, just where do you. What, what’s the step one here?

Zach Abramowitz [00:41:18]:
So listen, I think depending on your scale, it sometimes makes sense to hire an advisor, an AI Sherpa, right? So, you know, that’s, that’s obviously like where I come in for a lot of our clients. But I think this, the, you know, let’s put like the advisor side, you know, the advisor piece to the side for a bit. I think the most important thing first is to have a strong fundamental foundational understanding of how large language models work.

Zach Abramowitz [00:41:52]:
Right and I think that that can be done in a few ways. First of all, There are some online courses. This is something that we do when I, when I speak still today, I find that most attorneys don’t actually know how models are trained and how they work. And what I’ve also found is that those who do understand at a deep level, right, even under the hood, even with a little bit of math, they tend to have a much better user experience with the tools. They know what to expect. After you have that foundational understanding, I think you build on that through AI. Number one, asking AI about how AI works, I have to do this sometimes for myself. I’m like, oh, wait, can you explain positional encoding again? Because I thought I had it, but I didn’t, right? So I think that’s number one is just understanding how it works and then experimenting with AI and again, AI pilling yourself.

Zach Abramowitz [00:42:52]:
That is, that is the most, the most important thing that you can do is, is put on the goggles and realize, oh my gosh, like, this is, this is, we’re capable of a lot. I think the thing, the next thing is just continue further learning, right? So like I tell people, you know, there’s my podcast, but I also say, like, it’s probably worth it to listen to something general. The one I recommend is the AI Daily Brief. I think you listen to the AI Daily Brief once a week. You’re, you’re already ahead, you’re staying ahead of where the news is. You’re, you’re picking up sort of like high level trends. So I think continued, like, learning is, is another, like, very, very important piece of this. It’s just that the reason I say starting with like, real, not just like knowing how to use AI and which doc and where to attach a document and how to, you know, craft a prompt like that I think is less important.

Zach Abramowitz [00:43:49]:
I think that’s where most of the AI training has been. But I, when I see like the top attorneys and how they use AI, it’s almost always the case that they can explain. Attention is all you need. Vector embedding, they can explain, you know, agent harnesses and loops and they get it right. And I think that just having that like, foundation also makes you, like, unafraid of like, future learning. Because if you’re missing some, like, basic building blocks, everything, you could listen to all the podcasts in the world, it’s going to sound like gibberish.

Kyle Lawrence [00:44:26]:
That definitely makes sense. Well, with a few minutes left in today’s show, what can we expect for the rest of the year? I mean, normally I say, what does the next five years look like? But five years from now, I shudder to imagine what it’s all going to look like. So the rest of 2026, what do you see coming on the horizon? And if you can. And I have all next year.

Zach Abramowitz [00:44:43]:
So I have to, I have to think, I have to think about this because I’m getting asked to speak a bunch at the second half of this year. And I’ve spoken, you know, for I, I speak, you know, fairly often, as you mentioned in the intro. But a lot of people like want me to like give them the name of the title and topic for like a, for like a talk in November. And I’m like, are you kidding me? So what I, what I always say is like, if you really want a title for six months in advance, we’re just going to call the talk. Are we in an AI bubble? Because I know that AI could comb through a data room, do your entire like stock purchase agreement, draft the disclosure schedules, get all the documents signed, get everything ready, massage you, do your taxes, and we would still be asking, is AI in a bubble? Now that being said, here are, I think the stories that we’re going to be paying most attention to. Number one, I think people are realizing that AI isn’t just good at like the easy stuff so you can focus on the high level strategic work. AI is actually very good at the high level strategic work. It’s kind of sucky at a lot of the simple stuff.

Zach Abramowitz [00:45:55]:
So I think we’re going to see a move away from like more sort of mundane use cases. Like, can it draft my NDA? It’s like it can, but like, don’t you have an NDA template? Like, why do you, why do you need that? So I think we’re going to start seeing that. I think one of the ways you’re already seeing it manifest is in AI moving into the litigation space much more. So I think you’re starting to see a lot more attorneys who are taking the kind of core documents, not discovery, right, but like the core documents that got pulled out of discovery, putting them in, you know, either a Claude project or using one of the sort of specialty tools to do this and starting to craft their narrative, starting to like work on their strategy for trial. This to me is, is, is part of that move of thinking in terms of, about AI in terms of its like creative and strategic abilities. So I think we’re going to see the move there. I think you talked about before you asked me about like consumption and pricing model. So.

Zach Abramowitz [00:46:53]:
Right. Like Gora has moved to a consumption model and there was some serious backlash about this. The problem is we’re still in the experimental phase so actually you want to subs even if consumption makes more sense. By the way, I know certain companies that price have been pricing on consumption for a while and that’s been the right business model for them. Trial kit, which is one that specializes in and obviously tools for, for trial and discovery but being used by a huge number of lawyers in criminal defense and in family law now moving more into, into the civil space as well. They’ve been, they’ve always been priced on consumption and that was, that was the right model there in the big law world when you’re really trying to get people to experiment with AI and play around with it. If you’ve got a consumption model that is going to really, that’s really going to limit people’s ability to experiment. That almost feels like we’re going back to the days of like you know I remember when they trained us on like using Lexus and Westlaw.

Zach Abramowitz [00:47:57]:
It was like we got to be very careful about how you do your search because if you do more too many of these we’re going to like rack up a pretty serious bill for the clients. So I think that is going to be a story moving forward and I think as sort of an outgrowth of that is the story about post training. So I know there was a whole lot of attention has been paid to like the number of model providers that originally tried to fine tune the models and that was considered a kind of a big mistake, a waste of time. The story is that Bloomberg GPT which was a, was based on 3.5 and had been fine tuned on all of Bloomberg’s financial Data turned out 4.0 like outperformed Bloomberg GPT. So that was seen as a kind of like mistake. I think we could start seeing a, a return to post training. We saw it recently at base 44 and Harvey have been experimenting with post training models and the post training there is not only about getting the correct answer, it’s also about can we do this in the most efficient way so that sending an AI to look through the data room isn’t more expensive than sending a team of attorneys through which I think is also going to be a significant story. I think the, the story of buy or build your own will continue to be a very important one.

Zach Abramowitz [00:49:19]:
That’s partly an outgrowth of this I think open source also sort of an outgrowth of like the pricing question. And you know I say all the time like there was a period of time where music was free, it was called Napster and then it went away. We didn’t have Napster anymore. And then today music is free. It’s called Spotify or Apple Music. Is it, is it free? Free? No. You pay some amount of money, but if you compare what you get in an Apple Music subscription to what you would have gotten for spending 15 bucks at, you know, Sam Goody back in the day, like, you know, it’s, it’s, it’s totally, totally, totally different. So I think Consumer surplus.

Zach Abramowitz [00:49:54]:
Yeah, yeah, I think those are, those are going to be serious stories moving forward. I think the other thing, and this is one I’m yet to wrap my head around, but very important to pay attention to, was Alex Karp’s interview the other day on CNBC where he basically calls out the fact that AI is not having the intended impact it is at major, at big enterprises. The fact that big enterprises feel a kind of Faustian bargain. And like some supposed to give up all of our IP and know how to some forward deployed engineer who’s helping me like codify all of our knowledge into this AI model, Like, are we sure we want to do this? Like, like, you know, they’re like, yes, yes, go ahead. Implement AI. So I think that this is, this is a question and I’m not sure exactly what to make of it, but I do think that’s going to be a very, very important topic moving forward, which is are major enterprises really able to implement AI in a way that’s more than just giving their people access to chatbots, which by the way is pretty useful for those people, or will they be able to do something more than that? And how are they actually going to feel about turning over all of their data, all of their know how to the models and to the four deployed engineers that are trying to implement those models? That I think is going to be a huge question.

Kyle Lawrence [00:51:31]:
Wow, fascinating stuff, Zach. We could probably spend another six or seven hours today just talking about not just globally, but what our firm has been doing and how we’ve been harnessing this tech and rolling it out. Really fascinating stuff. I can’t wait to have you on again in six months when this entire world is completely changed. Really. Thank you for your time and for coming by. Zach Abramowitz from Killer Whale Strategies and Legally Disrupted Podcast. Really appreciate it and we’ll see you soon.

Zach Abramowitz [00:51:59]:
Thanks guys.

Moish Peltz [00:52:01]:
Thank you.