In this episode of AI and the Future of Law, Jen Leonard and Bridget McCormack welcome two leading voices in legal innovation: Jae Um (co-founder of Lumio) and Ilona Logvinova (Director of Practice Innovation at Cleary Gottlieb). Together, they dive deep into how AI is reshaping law firm operations, leadership culture, and talent development. From integrating AI agents into workflows to reevaluating law school curricula, this episode offers an honest, strategic look at how the profession must evolve.
Key Takeaways
- AI Tools in Daily Legal Workflows: Learn how leading legal professionals are integrating AI for notetaking, client communication, creative brainstorming, and task automation, boosting efficiency without compromising legal quality.
- Legal Innovation Through Strategic Acquisitions: Explore how firms like Cleary Gottlieb and AAA are modernizing by acquiring legal tech startups, embedding new cultures, and launching future-ready service models.
- Rethinking Legal Education for the AI Era: Discover why traditional law school curricula are falling behind and what legal educators, employers, and students must do to prepare for AI-native lawyering.
- The End of the Apprenticeship Model?: Understand why the decades-old legal training model may no longer serve lawyers or clients and how digital agents, AI simulations, and structured innovation programs offer a smarter path forward.
Final Thoughts
This episode challenges the status quo in legal education and firm operations, offering clear, actionable insights for legal professionals and institutional leaders. Whether you’re inside a law firm, law school, or legal tech company, this conversation reveals the strategic and cultural shifts necessary to thrive in an AI-driven era. Now is the time to rethink how we grow, train, and deliver in law.
Transcript
Introduction
Jen Leonard: Hi, everyone, and welcome to the next episode of AI and the Future of Law. I'm your co-host, Jen Leonard, founder of Creative Lawyers, joined as always by the fabulous Bridget McCormack, president and CEO of the American Arbitration Association. Bridget and I are thrilled to be joined today by two of our favorite thought leaders in legal: Jae Um, who is co-founder and Chief Growth Officer of Lumio, and Ilona Logvinova, Director of Practice Innovation at Cleary Gottlieb. Hello, everyone.
Ilona Logvinova: Hello! It’s so good to see you.
Jae Um: I’m so happy to be here.
Bridget McCormack: This is going to be so much fun, you guys. Thanks so much for doing this. It’s going to be a really, really fun conversation.
Jen Leonard: This is going to be a party, I can already tell. And we are going to kick it off with our AI Aha! segment (co-branded — tip of the hat to Ilona’s colleague Megan McMillan at Cleary). AI Aha! is our segment where we hope to inspire people in legal to use AI in lots of different ways in their lives, personally and professionally.
AI Aha! Moments
So we want to hear how each of you have been using AI. Let’s kick it off with Jae. What have you been using it for in a way that you find particularly delightful these days, Jae?
Jae Um: I have a super fun one. Are you ready? So it is a custom agent with custom instructions — not very long, but you know how everyone’s been complaining about how GPT’s writing style is so fake and it sounds like AI? My instructions are all about tone and voice.
So I created basically a reverse Key & Peele “Obama’s Angry Translator”. Remember Luther, Obama’s anger translator? It’s a reverse Luther. I type in super angry emails, and then it just spits back to me lovely, gracious, patient emails.
Jen Leonard: My gosh, that’s amazing. I love that! Ilona, what about you?
Ilona Logvinova: I’m really inspired by that. Mine’s going to be a lot simpler, but it’s so effective. So on the professional side, right, I’m constantly on calls. A lot of my day is taken up with back-to-back calls. (Those of you who know me know that I start my calls five minutes after the half hour or hour, to give myself that little buffer.)
I take a lot of notes during my calls — I’m always in OneNote, taking notes. And I’m constantly, constantly now using Labora, which is our partner platform internally (our all-purpose platform). I take my notes and I prompt the AI to synthesize them, pull out action items or an executive summary. In that five-minute buffer between my calls, I can send those notes out incredibly quickly, right after the call, and then it’s done. So I don’t have that running to-do list through the day — I have my instant takeaways. I’ll prompt it several times: “What are the takeaways? Did I miss anything? Is there anything inferred that we should make sure to do?” Just generating insights based on the notes I’ve taken. It’s a closed workflow loop that’s beautifully efficient. I highly recommend it.
On the personal side, my daughter’s nine years old, and she’s very creative and always drawing things. We spend a lot of time doing art projects. She’s always asking what she should draw or what scene to create, right? So I’m always asking ChatGPT what we should draw — kind of creating with ChatGPT and figuring out what we can put pen to paper on. And that’s been a lot of fun. She’s like, “Mom, stop using ChatGPT, we should just talk!” And I’m like, “No, but this helps us be more creative, right? It’s a creative agent.” Now she’s gotten really excited by it, too.
Jen Leonard: I love that. I also have a nine-year-old daughter, and she also loves to ask ChatGPT for inspiration and then go draw, too.
Bridget McCormack: Those are both amazing examples. I need to go back to both, though. Jae, did you create that agent on OpenAI’s platform?
Jae Um: Yeah. And actually, I used GPT to help me with the tone and voice analysis of the outputs I wanted. I gave it some samples of emails that I thought were particularly kind and gracious, patient and still concise, and then it took me down a whole rabbit hole of like, what are the components of voice? I then asked it to help me tighten the system instructions.
Bridget McCormack: Awesome. And Ilona, why do you take your own notes instead of just letting an AI take notes? Because you know you’ll take better notes?
Ilona Logvinova: I do use it, but I find it’s my own takeaways. It’s not the whole meeting, right? We all know how this works — you’re taking the strategic parts away from the conversation. And my goal is that there should never be a meeting without action items, from my perspective. If there’s not an action item drawn from the notes, that’s a problem. And doing it this way helps me elevate that insight — to evaluate, you know, was this an effective meeting? Did this make sense? Right. Could this have been a Teams message or an email?
Bridget McCormack: That’s super interesting, because aren’t we all trying to figure out which of our meetings can be emails or Teams messages?
Jae Um: I love that, though. Sometimes I see a lot of people — even people who are very into AI — not being intentional about who should take the pen on the first draft of anything. I think, you know, it’s not a monolithic situation where one or the other is always the right answer. I think it’s application-specific. And I really like the idea of us doing the first run of the notes, because I’ve generally been unimpressed with call-recording-transcript–driven notes. You end up spending more time weeding out mischaracterizations or irrelevant bits — sometimes the AI doesn’t actually understand what you’re talking about. That’s brilliant. I’m going to try that.
Jen Leonard: I really liked that recent One Useful Thing blog post that Ethan Mollick put out, the “think first, meet first, write first idea”. I was playing around on a writing project this summer, and just out of curiosity I asked both Gemini and ChatGPT to write the article I was working on. I gave them prompts and context (I think I might have even done some research first), and I was not really impressed with the first drafts. I was curious to see whether it would be easier to edit a first draft than to write it myself, and I ultimately found it much more frustrating to write that way.
So, Ilona, I can totally see the benefit of taking notes yourself and then capturing the takeaways. And Jae — I could actually see the benefit of using Ilona’s emails to train your agent, so it learns to write the world’s loveliest emails. I wish I could write like her, with her tone… The world should capture that!
Jae Um: (laughs) I can confirm: she is the world’s most gracious email writer. I tried the AI-first-draft approach to writing as well, because, you know, I still write quite a bit. The thing I’ve found most helpful, actually — what I’ve been doing recently when I have to draft stuff — is to get AI (whether it’s GPT or Claude) to interview me.
I just give it the topic and the audience, and then I act like a reporter, or like a student, or like someone in the audience. I tell the AI, “Ask me one question at a time,” or I say, “Give me five questions.” Then, you know, it’s more of a brainstorming exercise. I find that, just like I was saying about being intentional about who takes the pen on the first draft — sometimes when we jump straight to the final deliverable, the output you’re trying to create, you don’t get the best results working with AI. I think being a little open-minded about the process gets really good results for me.
Jen Leonard: I love that. Do you always type with it, or do you ever use voice mode for the interview?
Jae Um: Ed Sohn — who’s our head of product and one of my best friends — has been reminding me to use voice mode for like a year and a half now. And for whatever reason… I’m a typer. I’m a typist. That’s just me.
Bridget McCormack: Even when you have your phone — like, if you’re walking around and you want to continue the conversation on your phone — you’re typing?
Jae Um: So one of my work flex moves was sending these perfectly formatted emails sent from my iPhone. And you guys know I blow up your phones all day with texts — I’m a furious texter. So even on my phone, typing is what’s most natural for me. I tried voice mode, but I think I just don’t like waiting for it to speak back to me, and I don’t love recording voice notes myself.
Jen Leonard: It’s interesting. I love the multi-modality of it, because one thing I’ve realized since I don’t work in an organization anymore is that I really crave conversations like this with really smart people. I learn as I talk out loud. So I love voice mode, because I really do think I make connections I wouldn’t make through typing. I think I’m going to try your approach — do an interview with voice mode — and see what happens.
Jae Um: That makes a ton of sense. Actually, I am an internal processor — I think by writing, like I always have in my career — and Ed’s very much an external processor. External processors love voice mode. But I love multimodal AI. What I love are images and photos. Like, I love taking pictures of things when I’m walking around. I’ve tried the “take a picture of what’s in my pantry and fridge and ask what I can make” thing. And I’ll scribble ideas on a notepad and then take pictures of it. I love that. But voice mode is not my jam.
Jen Leonard: That’s why it’s cool — AI is like a Swiss Army knife of tools.
Well, thank you so much for your Aha!’s. These were fantastic.
Culture, Strategy, and Innovation That Actually Works
Jen Leonard: Moving into our conversation about the profession… We’re going to turn to Bridget first to kick us off on the future of the legal industry and all the change that’s ahead of us because of AI.
As our audience knows by now, Bridget has been leading an AI transformation at the AAA for three years now — which is amazing and hard to believe, I’m sure. And you’ve talked at different times about the innovator’s dilemma and the challenges of driving change within a highly successful, legacy organization. Could you kick us off and talk a little bit about the challenges and the value of growing organically versus the value of inorganic growth — bringing in different components to allow you to grow in new ways?
Bridget McCormack: Yeah, absolutely. (Although I’m most interested in hearing Jae’s and Ilona’s reactions to all of it, and their advice and counsel, because we’re so lucky to have them here!) So I won’t talk for too long. I think Jae and Ilona know some of this story, but: the AAA is going to turn 100 years old in 2026. And it’s a well-run business that provides dispute resolution services around the world — a lot of them. We do over half a million cases every year. We have an awful lot of expertise and a lot of information that isn’t really found anywhere else.
So it seemed to me that — like every part of the legal profession — this part is pretty ripe for disruption by AI. We had to figure out what that was going to look like in an organization where most people have worked here for a few decades. (That longevity is part of why we’re successful: we have experts who have been around for a long time doing excellent work.) But, like in any legacy successful organization, it’s really hard to pivot to do something that might disrupt what you’re doing today. You need a real plan for that, and change management is part of it.
In my view, inorganic growth had to be part of the solution. It’s not the whole solution, but it had to be part of it. So, I don’t know if you know this story: the AAA had its first acquisition in our 100-year history. Last year we acquired a little Silicon Valley startup that does online dispute resolution (which is a big term that can mean a lot of different things). The point is, they were operating in a space that we were not yet operating in.
Which isn’t to say that the talented people I work with couldn’t have done it — of course they could. But by bringing in this new business (it remains a for-profit subsidiary of the nonprofit parent), they could operate as a tiger team in building and standing up new services for new markets. And then we could learn from each other. There were all kinds of synergies, and we could deliver some of the services and products of the future in more experimental and quick ways — get quick feedback from users in the market.
At the same time, we were doing a pretty significant transformation with the core team (as Jen knows, and has told the story for us beautifully). In my view, that’s gone maybe better faster because we introduced a new team with a new culture. We were able to learn from one another and also support one another in places where each of us was not as strong.
So there’s a lot more I could say about it, but I know Ilona — and Cleary — have some experience with this as well. I’d love to hear your take on it, Ilona.
Ilona Logvinova: A little bit of experience, one could say. (laughs) But Bridget, I love hearing your story. It’s so inspiring, because it was such a great idea, right? The idea that you’re taking an organization that has done something extremely well — you are the best at what you do in your industry, right? — and you really looked to modernize it.
What I think is so interesting and important about inorganic growth is not just bringing new capabilities into an organization, but it’s the opportunity to bring a whole new culture, right? A whole new way of working, a new operating system internally into an organization.
And of course, once you do that, you are giving the acquiring organization a chance to learn from that new entity, to learn the systems and the ways and the approaches of the new entity. We can talk at length about integrations more broadly when it comes to M&A, right and where they fail and where they do well, what that looks like.
But in my mind — and I’ve seen a number of integrations in my professional experience now — the integrations that are truly successful are the ones where the parent organization (the acquirer) really takes on the culture of the entity they’re acquiring, right? They invite them in and they learn from the experience.
And it seems, from what I’ve heard from you, that you’ve really done that — you’ve allowed that transformation to take shape pretty organically through inorganic growth. We have certainly tried to do that with the acquisition of Springbok AI, which happened about four months ago (in March of this year).
And the idea was really a similar perspective, right? We have a craft that we do well at Cleary — we have expertise across industries, across regions. But the legal profession is changing, right? That’s what we’re all essentially talking about here today, and that’s what you and Jen talk about on this podcast. Understanding that to be true, what is it that will help us change, right? What will help us transform?
So the vision for bringing in Springbok was really to embed data analytics into the way that we’re working — to help us work differently and really enhance our client service from the inside. Right from the start of the workflow to that client delivery.
So it’s been a really exciting journey, and it’s been a fantastic integration, really, because I give a lot of credit to the firm and to our leadership team for inviting this new culture into our culture — into a culture that was already great — but recognizing that it’s not that we need to change, right, it’s that we have an opportunity to transform. I think those are two different things. And if you get the latter right, you can really benefit from that invitation to understand and align with a new culture that’s brought in.
Jen Leonard: Amazing. And Jae, you have the opportunity to have a little bit of a bird’s-eye view of what’s happening all across the industry. What are your takeaways about combining different forces for growth? What works well? What doesn’t work well? What are you seeing?
Jae Um: Yeah. I think that one of the shifts in the industry that I’m seeing — that I really like — is a change in tonality and a change in posture and attitude toward innovation and new things. I think that there is much, much broader recognition now than maybe ten, fifteen years ago of the benefits and the opportunities that are afforded by new ways of thinking — welcoming new and different capabilities into the organization.
And, you know, I’ll take this back quite a bit to my time at Seyfarth. Seyfarth was very early in doing a lot of this thinking. Seyfarth had subsidiaries in adjacent businesses. The Seyfarth Shaw at Work program was quite an older business, actually — I think they started that in the late ’90s — and it was the training arm, which made a lot of sense given the labor and employment practice group of the firm. And the SeyfarthLean Consulting initiative, which was ahead of its time, really, because, you know, it was designed to help corporate clients make the jump to digital business and explore different ways to engage with their providers, as well as — you know — service model innovation for the firm proper.
So I think when I think back to some of the work, the culture work and the communications work that I did with Stephen Poor — I was so lucky to learn from his years of experience and wisdom — the respect for what exists. I think it’s very important that innovators bring that to the table.
And I think it is really a new idea in the world, because if you go back to the intellectual history of innovation: the older idea, like from the 1940s, is creative destruction, right? That’s what Schumpeter called innovation — that new things were always going to happen in the world, and that they would destroy or displace old things. I think that’s an adversarial dynamic that maybe, you know, scaled to the sexy magic of Silicon Valley and all this talk of moving fast and breaking things, right? I never thought that that was appropriate for the law. We’re talking about law firms that literally created the regulatory legal framework of this country. There are firms here that are older than the United States. And there are firms that, you know, kind of lay claim to really important legal innovations — like the holding company, like the bankruptcy practice, like the trust indenture. There were lawyers and law firms that helped build all of that out.
Right? So I think the people trying to improve the practice of law really have to come in with some recognition of what has happened in the last — not just 100 years, 150 years, 200 years. But it’s also that the people who have been there doing amazing work understand that new and different capabilities are sometimes necessary and sometimes they are transformative, given how the firm can discharge its core purpose.
Right. And then — turning everyone’s attention to why the organization exists, right? Like, who are we serving? What are we meant to do? What are we all trying to do here, and what are the best available ways for us to do that? How does a new acquisition, or a new group, or a new capability change our ability to create value for our stakeholders, for our clients? And I think kind of giving people a shared purpose is super important.
And now I think, you know, that innovation has certainly moved upmarket. There are many people in innovation leadership roles at some of the most profitable, whitest-of-white-shoe firms here in New York and certainly in London also. I think that culture work is much, much further along than, you know, what I saw in the industry 15 years ago. In terms of acquisitions, I think that the strategy part of it needs more thinking, though. I think that it’s not just an emotional or cultural distinction — whether you’re trying to displace the way things are done, or to enhance them, or to create entirely new ways of doing old things, or are you going to try to do new things, new value propositions that you haven’t tackled before? I think there are strategy implications as to whether you’re going after process displacement or, you know, kind of outcome enhancement for the client, or creating new service lines and new service delivery models for existing revenue streams, or opening up new revenue streams to address new needs that you haven’t served before.
I think that level of strategic thinking seems to be lagging some of the investments, which is not surprising. But I think that in the next five years we’re going to see, you know, a lot of old, established legal businesses really lean into that thinking about how do we monetize these investments?
How do we reach that — not only problem-solution fit, but product-market fit, right? Who are we trying to serve with what offering? So I think that’s probably what’s coming next.
Jen Leonard: Can I ask you a follow-up on that? This is sort of stemming from advice that you gave me as a solopreneur, but at scale. You told me when I started my business, you said, “People will tell you you need to focus. Don’t be swayed by that — experiment and see what you enjoy doing, and also what is profitable.” Do you think law firms need to focus on one of those different strategies that you mentioned, or can they experiment with different things and try to do many things at once?
Jae Um: So I’m going to start — just as a joke — with everyone’s favorite lawyerly answer: It depends. But I will follow up! I think that it’s an iterative process.
So where I think firms need a focus lens first is understanding their current assets and liabilities. And I mean strategic assets, strategic liabilities. And the reason I think that is more important for them than, you know, for you and me, Jen, when we were solopreneurs or founders or entrepreneurs starting new things, is that they are encumbered by an entire history of accumulated choices. They do have a lot of stakeholders. You know, it’s not just a business. Absolutely, some of these most profitable firms in the world — they already run like businesses (like, the idea that they don’t… I reject that entirely!). But there are also social and political entities, absolutely, with a lot of stakeholders who have a lot of stake in the system. Right. So I think that it’s important for them to take stock of their current position: Where are they strong? Where could they be better? Where could they maybe pull back past investments and redirect that to, you know, new directions? I think the focus mindset has to start with where they are now — the organization’s current reality.
I think when it comes to the organization’s future aspirations, it’s extremely important for leaders to help people kind of switch their mental context — their emotional context — to exploration, so that people can learn about emergent horizons, new possibilities, and create new thinking, new framing, new ideas. Right?
Because — what’s the Einstein quote? — we can’t solve problems with the same mindset we had when we created them, right? If you want people to be innovative, if you want people to have new ideas, sometimes you have to take them out of the office, put them in a new physical space even for a day. You have to expose them to new ideas and new stimuli, right? So I think in that opportunity scanning is really important — to be exploratory. But the synthesis of the two, I think, is immensely important, because the question really becomes: What are the new possibilities where we are better positioned than other organizations to go after those? You know, because opportunity doesn’t exist in a vacuum. You really do have to take a disciplined, focused look at what your assets and advantages are, and then evaluate the whole, you know, kind of universe of possibilities against that lens. Right? Which are the right, best opportunities for us that we really can’t miss — like, we have to take a swing? Which opportunities might be a reach, which are truly distractions that are going to really hurt our chances of succeeding at these other things where we are well positioned — where the strategic advantages are mission-critical to our future advantage.
Ilona Logvinova: I love all of that. As you were talking, Jae, I was thinking about how important it is to think about the value proposition of what we’re delivering. Right? Jen, I think I’ve heard you say this first, but the law — as an operating system — has not had a refresh in quite a long time. Right?
And one of my favorite quotes — I think it was in an Ethan Hawke TED Talk — was about how, you know, creativity is not a luxury, it’s actually sustenance. A beautiful TED Talk; I highly recommend it if you haven’t seen it. But I think it’s the same now for our profession, right? And for firms who are looking to stay ahead of the curve, ahead of, you know, the changing industry, we need to understand what our clients are doing, how they are working. How does that inform our delivery model? Because exactly to Jae’s point — right — nothing exists in a vacuum. And so when we’re thinking about delivering value, it has to be derived from: What is our best and highest form of advisory, and what does that advisory look like?
And it probably won’t look the same — or it ought not look the same — as it’s looked for the last number of decades. Because we don’t have that sort of long-form attention anymore, right? People don’t want to read 40-page memos. People want data visualizations. They want interactivity, right? They want immediate, accessible bullet-point information. And that’s a very different delivery model than what we’ve been used to in the legal profession. And that takes different tools. It takes a refreshed workflow, right? It takes a different type of training. Maybe the pricing model is different, right? Maybe there’s tech on the front end — I certainly hope there’s tech on the back end.
Right. But all of this means that the way that we’re doing it is different. And I think the really exciting piece right now is figuring out what that “different” means, and what are its component parts, and is it modularized? And how do you assemble it, and how do you make it fit for purpose across your practices and your regions so that it best serves your clients?
Bridget McCormack: I mean, you guys are all experts in this, but the barriers to doing all of that for busy lawyers are infinity, right? Lawyers are too busy to make time for that. They’re trained to think that they can’t experiment with creativity, right? That’s actually part of the training — part of what they’re taught. They’re doing quite well right now, right. Even though everybody’s been talking about how law is ripe for disruption with AI, it looks like profits are okay in the big firms. Right. So there’s a reason for them to be skeptical. And there are lots of reasons for them not to make time or space for the kinds of work you two are suggesting they need to do.
Are you seeing them do it? Is anything changing yet? Or do you think this is all still to come — and there’ll be, you know, some wake-up calls that force it?
Ilona Logvinova: I can kick us off, maybe, with a very short snippet. You know, without giving away too much, I’m starting to see a change. I don’t think there’s a sea change just yet, but I’ll say, you know, as an innovation leader — as someone who’s focused on this — this is very much something I’m thinking about constantly. And this is the vision state, right? This is on the vision board in terms of what we’re doing and what we’re setting up for, so that we can deliver differently. It’s the why. And even if it’s not fully here yet, that is what we’re driving towards.
Jae Um: I have a balanced answer. Some of it’s going to be depressing. I think it depends what you mean by “they.” Like, who is “they,” right? There’s a lot of ethos that goes into innovation forums and among the kind of people who dedicate their entire lives to innovation. I think they do bring a very strong ethos to the challenge. But I think when you look at markets, the pace of adoption of innovations is quite practical.
I think that commercial enterprises adopt innovations when it makes financial sense to do so. And so I think that pockets of the industry that have been under greater financial pressure — in terms of rates, pricing, margins — they went first. I think we can clearly see that in some of the ways that service models have changed for legal work that I would say is like “core-minus,” closer to commoditization. You certainly see that… if you go back to the dot-com bubble era, right after the bubble burst, things like offshoring work to lower-cost regions of the world were very new. And that’s standard practice now, I think, for global organizations. If you look at some of the back-end processes that go into e-discovery, that was very new — and we had to have the Sedona Conference, and we had to have some lawyers (very few lawyers) focus really hard on that, right. And once they figured out a scalable methodology, businesses just sprang into being, and those founders ran very hard to make e-discovery a huge segment.
LPO is the same, ALSP is the same, a lot of the work that you see in contracting now — does that affect the ivory-tower profitability of some of the oldest firms in the world? Not yet. But, you know, here’s where Christensen’s model of the innovator’s dilemma always holds true, right? Innovation does start in the lower end of the market, and then it does move upmarket. I think we’re seeing that upmarket move right now. As somebody who paid for college by being, you know, a secretary and a litigation paralegal, I remember — I am somebody who has physically stamped evidence. So I know that change does happen in legal practice. It may be, though — and this is the depressing part — that when change is broadly adopted finally by the majority of the market, it’s no longer exciting, because all the novelty has worn off. And the people who had the idea, people who believed in it when it was new, they're so worn down by the change-management battles that the light has just gone out of their eyes. But that’s the nature of it — new things are thrilling, but not all the thrills are good. And then by the time everything feels safe enough to adopt, it’s not exciting anymore. And so I think that hype curve has an emotional arc to it.
Now I’m going to go back to your actual question, you know: Who should be doing that work? I think it is not fair to ask, you know, associates who have 1,800- or 2,200-hour billable requirements to explore in a vacuum, in a directionless way. I think that is, frankly, nonsense. They can’t. I think that’s why it’s so important for leaders to create parameters, create spaces, so that people can have mental space, emotional space, and be in the right headspace for exploration. That’s why I think programs are really important. I think that’s why the correct — and I do think correct, because there are correct and incorrect — configurations for innovation are important. I think choosing the right configuration is really important to maintain today’s success and to invest in tomorrow’s success. Right? So culture, investment, programming, operations, infrastructure — I think these are all components of innovation leadership.
And it doesn’t rest just with people like me, who are consultants giving advice, you know, or founders who are creating new things (as I am doing now). And it doesn’t rest with people like Ilona, you know, inside the organization driving that change. Ultimate accountability rests with the chairs, the managing partners, the chief operating officers — because they are accountable for how the resources are allocated today.
So I think they really need to lean into the challenge of evaluating ideas that they can get from many places, and then, you know, creating the configuration that works.
Driving Change in the Legal Profession
Jen Leonard: Can we talk a little bit about change and learning? Maybe, Jae, you could describe “Meng” for us and explain what pandas have to do with social learning — and what social learning has to do with AI. Then we can have a conversation about maybe why just dropping licenses for a new AI product into the hands of lawyers is not the best way to effectuate adoption and change the way that we work.
Jae Um: So the picture is of Meng breaking off a piece of bamboo — because that’s what pandas eat. And Meng is a panda. (Makes a hilarious, adorable face every time he does it.) And he makes that face because Meng is an orphan who was raised in a zoo with human caretakers. He didn’t have panda parents to teach him how to eat. He learned that from human caretakers. And the reason Meng still makes that face every time is because he watched his human caretakers break up the bamboo.
I thought it was such a cute story — and it has a lot of lessons for us. It actually made me reflect a lot on the nature of learning. Ed Sohn introduced me to Social Physics a few years ago. It’s a great book by Alex Pentland, I think — highly recommended — about the idea of social learning as a construct.
So I think Meng’s story reminds us that when you go to the most primal, most basic part of our brains, learning begins as a deeply embodied social process and is so trust-based and context-dependent, right? Babies start by imitating — mimicry. They express themselves by copying what adults do. As they start to mature, that’s when they start connecting, you know, effort to intention and goals. Right. And then that’s when they start emulating the people they trust (typically adults in their lives).
And I think that when we kind of connect to the nature of learning — how much trust matters, how much context matters — and we think about maybe how we have been training associates for over a hundred years, how we expect people to learn in the workplace as fully grown, formed adults with their own goals and intentions… How well are we connecting the learning that we expect lawyers to do with the goals they have, with the intentions they have, with the constraints they have? And then how are we thinking about the job relevance — the immediate job relevance — of the things we’re asking them to learn? And how are we thinking about the social compacts that, you know, kind of form the boundaries of the trust they have with the partners and the partnership as a whole? Right. I think there are lots of lessons to be learned from Meng. So I hope everybody can see a picture of Meng! (laughs)
Jen Leonard: Ilona, anything that resonates with you from the work that you’re doing to sort of drive AI adoption and learning? Or anything just generally about the profession and the future of change?
Ilona Logvinova: I mean, so much to say. First of all, I love Jae’s framing. I’m going to check out that book. I think the broader theme for me is intentionality. Right? I think what’s happened in the legal profession, historically and through now, is that we sort of have this apprenticeship model via osmosis. Right? So people are learning kind of through the process of hopefully picking up things that they need to emulate. But there hasn’t been — a true apprenticeship model in a very long time. Right. Unless you get very lucky and you’re an outlier. But mostly people are in their offices, in their own spaces. Right now, increasingly, we are moving to digital.
And so there is this sense of, you know, people are in separate spaces, and there’s this perspective that it may be harder to learn or harder to, you know, have that apprenticeship model exist. And I actually think that it’s quite the opposite.
I think that if that’s the perspective and they’re not seeing all of the opportunity… Because I think the hybridization that we’re seeing now, and the transition that we’re constantly doing from analog to digital in terms of how we’re collaborating, how we’re working — right — moving from physical spaces to Zoom and back and forth, and having hybrid meetings, there is so much more of an opportunity to do more intentional digital learning.
Right. And there’s great work that Megan Ma is doing at Stanford Law, through her lab, in terms of agentic approaches to education and agentic simulators when it comes to actual learning, right, on the job — and how lawyers can learn to craft arguments and negotiate and, you know, write responses and emulate a certain persona or a style of lawyering. There is just so much opportunity to inject a different level of intentionality into legal education curricula that we have not done in a very long time, and also embed tech into that learning process. Because now lawyers will have to understand how to interact with technology, right? You will have to learn to talk with AI and to AI, and understand its capability and how it can be an extension of you.
And of course, you know, we haven’t yet brought digital agents into the conversation, but agents will be everywhere. And so we should all presume that going forward, we will all be supervisors of digital agents for sure. Right. And maybe other people too. But even at the more junior levels — on the most junior levels of our profession (of most professions now, right, or professional services at least) — you will be supervising digital agents, in the best-case scenario, because they will be doing vertical tasks in your workflows.
And so what does that mean for training, right? So I think that’s kind of a big answer to what is a very big question. But the overall punch line for me is that we have to rethink how we’re approaching this, because the process itself is changing so much, and so much of it has that opportunity for digitization, right, and the tech embedding and the learning that really just can be so much more exciting.
Bridget McCormack: It’s a lot of things kind of lining up at one time, right? Because, right — the hybridization of our workforces and our collaborations is one factor.
And — you didn’t say this exactly, but I’ll say it — I don’t think that the so-called apprenticeship model that we’ve kind of lived on for many decades now (of just assuming if junior lawyers did the same task over and over and over again — sometimes with very little feedback, sometimes with great feedback depending on who you got paired with, but sometimes you got really unlucky and got no feedback — but somehow we assume that, I don’t know, seven years later, by just doing those things over and over again, all of a sudden you’d have wisdom and you’d be able to be a strategic advisor)… frankly, that’s nuts. That wasn’t really the best model.
And so, in a way, it’s great that the new ways of working are forcing us to rethink that. But it’s not only that, right. It’s the fact that the business model is clearly going to change because of technology as well. And so that might be the faster or more urgent motivator for some stakeholders in legal organizations to say, “Hey, maybe we need to rethink how we can take a really smart, you know, recently graduated lawyer and turn her into a strategist. Because probably that big client of ours isn’t going to keep paying us to let her do that rote thing for seven years in a row.”
So a lot of things coming together at once. And you guys know the market better than I do, but I assume the second factor (the business model) might be more important than the first (the work model). There are some others as well…
Do law schools have a role here? I mean, I know you guys are maybe not focused on legal education, but Jen and I think about it quite a bit. Do you think that this is a moment where stakeholders across the profession — who usually stay in their own corners and don’t talk to one another — might think, “Hey, this is a moment where we need to rethink how we do all of this”? Any thoughts about a role for law schools or legal education in what’s coming?
Ilona Logvinova: I love this question. I think this is an opportunity for the ecosystem, right? For the law schools, for the future firms who are hiring, for those who are thinking about this more broadly (or advisors, or even undergrad). Right. So I was actually talking to someone at Cleary who’s about to go to law school. We were having a session before she heads out, and I was asking her if she knows whether her law school will be using legal tech, right. Will they be using legal AI?
And I think law schools should make sure that they’re incorporating AI into their way of working, right — the way that they would offer different tools like Lexis and Westlaw. Why not have licenses to, you know, platform-based tools that students can use? Because you would then understand things much more quickly, right? This is a new tool that you add to your toolbelt when you’re taking exams or when you’re, you know, doing X, Y, Z. And the idea is, you will have to embed this as a lawyer. And so why not start earlier on so that you can sharpen your own thinking and get yourself ready for that next step and make it less theoretical? Make it more practical, right? Understand how to do the right sort of prompting, because legal engineering will become more and more important. We all understand that prompting will become less onerous, right — in the sense that the AI will get better. And it’ll be less specific. But legal engineering — meaning injecting legal expertise into the questions that you’re asking and carefully crafting that so that you get the output you’re seeking, the expert output — that will be really important.
Jae Um: I have harsh thoughts about law schools. Ilona is absolutely right — I think this is a moment for the ecosystem. And in that, you know, I give full-throated support to what she’s saying, and I completely agree. But, as usual, I’m gonna bring kind of an economics/market view to the question. And I would tell you that, from what I’ve seen in ten years of work with law firm leaders, and in observing the economics of legal services in the enterprise market: I’ll tell you that law schools are no longer putting out a good product. And I’m talking about the product they offer to the students who are paying tuition.
When you think about the price point of a JD in the United States in particular, and then you kind of look at all of the systemic pressures that have been placed on the Lancasterian model of education and The Cravath System— and how much has changed in the world… like, one data point: last year, 360 billion emails were sent per day, per day in the world. You know, so when you think about the volume of information that is swirling around businesses, what that does to the nature of lawyering for, you know, some of the world’s most complex businesses. And we think about, kind of the long-standing notion that you spend three years learning to think like a lawyer and the firms are supposed to train you… I’m not sure that logic is going to hold anymore.
So when I think about the opportunities for law schools, I think there is room for, you know, shuffling of the established order. What I would tell most managing partners is: in the next five years, if you have never hired outside of the T14, I think it’s time to start looking beyond the T14. I think that it is time to start asking about the curriculum at some of these law schools in a systematic way. And I would be willing to go on the record and say my prediction for the next ten years is that it’s going to be a buyer’s market for talent. And so the balance of power already rests with the firms — you look at the most profitable segment of the Am Law 100… They have the power anyway. But that power is going to shift more and more toward the employer in the next ten years. And I think that there is really an opportunity for deans of law schools to rethink the curriculum, rethink the value proposition — going back to our earlier conversation — that they’re offering to law students.
I think that it’s time for law students also to be looking beyond the vault guide and the T14. I think that — that is also a platform business, right? Law schools connect students, turn them into JDs, connect them to employers. So I think that, you know, change has to happen in perception on both sides of that platform. But, you know, not to pile on law schools… I think the same about most graduate education. I think most graduate schools in the United States are no longer putting out a good product, and business schools certainly are trying — because they have more pathways to offer kind of topical coverage of new things, and business schools are always thinking about that. And I know medical schools as well have to keep up with advances in scientific research and new biotech, new pharma and all that. And law schools actually have fallen behind. If you look at the kind of the highest-paid professions, right… I think it’s time for that to change.
Jen Leonard: The other difference there, too — especially in the medical context — is there is almost a requirement that you be in an operating room or be making rounds to be teaching medical students, as opposed to teaching law, where there is a real divide between practice and academia. But I will say, I don’t know whether it’s a positive sign or not: there are more first-time law school deans than ever before coming into their positions this September. I think it’s a sign of how difficult the job has become, which creates opportunities for more new deans. I think there’s also the shortest average tenure on record for being a law school dean — I think now it’s under four years — which I think is a clear sign as to how difficult the job has become.
But I also think — just going back to Ilona’s point about how law firms should be driven by what the client’s goals are, and that that should be paramount — so I think Anthropic did a report about which undergrad majors are using AI the most, and it’s STEM majors. And I think if you look at how law schools are approaching the use of AI — how many are prohibiting the use of AI — I think if you spell that out and think about how client-centric you can be as a counselor to your STEM clients, eventually it’s just a nonsensical approach to being able to keep up with what your clients will expect from you.
So I think it just underscores your point, Jae — being able to look under the hood of the curriculum, and having the market reward schools that are keeping pace.
But sadly, we are coming up on the end of our conversation. So I’m going to toss it to Bridget, who has a really pressing question that I know requires society’s attention. It’s urgent, and we all need to be focused on it. And I know that we need the world’s best minds to be thinking about this. So I’m going to toss it to Bridget to set us up…
Bridget McCormack: And the world’s best minds are here. So — lucky for the world! (laughs) We assembled the best minds: Ilona and Jae showed up. So we get to answer the question. I want to put a pin in something for a future episode — you guys are going to have to come back, because I want to talk more about that ecosystem moment question. I want us to think about (for when we get together again): Who could be the convener of all the stakeholders? We talked about two stakeholders primarily — law schools and law firms — but what about the regulators, the judiciary, the public (who never gets a voice in how law is effectuated)? It is this ecosystem moment. And I wish we had a convener who might say, “Hey, this is a great opportunity for us to completely rethink how we approach our profession.” So, pin in that, because that’s an entire episode!
But since I have you… there is a really difficult legal question — you know, kind of a statutory interpretation question (not really a statute) — that divided the appellate Twitter community for a long time (when that was a thing… not so much a thing anymore). And the question is: Is a hot dog a sandwich?
I mean, think about it. Use whatever methods of, you know, statutory construction you favor, and give me your best answer.
Jae Um: Ilona, do you have a firm position?
Ilona Logvinova: I do, I do.
Bridget McCormack: Okay. I have a firm position as well. You go first.
Ilona Logvinova: I think it has to be, right? Because it is something between two slices of bread. So I think a hot dog is a type of sandwich, necessarily.
Jae Um: Okay, I’m going to take the other side. I think it’s not a sandwich, because a sandwich is stuff between two slices of bread, and the bun is hinged — like a taco. So I think it is a completely different category. I think they both belong to the “handheld meal” category. But look: it’s one piece of bread!
Bridget McCormack: (laughing) This is exactly the divide among the smartest appellate lawyers in the country, at the fanciest appellate firms. You guys just hit on both sides of the argument. You can keep thinking about it… Jen, do you have a vote?
Jen Leonard: Wow. I mean, I had not put enough thought into it, and now two of the smartest people I know are split! But you know what — on a note of optimism for lawyers, I just feel like there are so many questions that will keep lawyers busy for all time, and this is just the perfect example, right?
Jae Um: Right. And next time we have to look at that dress and see if we think it’s white or it’s blue.
Jen Leonard: Oh my gosh
Bridget McCormack: I am on record as taking the position that a hot dog is a sandwich. I came down where Ilona did. Jae identified the one weakness in the argument, but I don’t understand how something that’s in between two sides of bread isn’t a sandwich.
Jen Leonard: Wow. This was such a party — as I knew that it would be. And Bridget said we have to have you back for a future episode to imagine a stakeholder convening to try to bring together all these disjointed groups that should be thinking about the future of legal education in particular.
But for now, thank you so much for sharing your wisdom and thoughts about a changing industry. We look forward to seeing our audience on the next edition of AI and the Future of Law. Until then, stay well.