Subscribe
Share
Share
Embed
A RopesTalk podcast series highlighting Ropes & Gray’s technology attorneys, their respective practice areas and how the ever-changing tech landscape impacts their clients.
Melissa Bender: Hi, I’m Melissa Bender, a partner in the asset management group at Ropes & Gray, located in our San Francisco office. I want to welcome everyone to the latest edition of the R&G Tech Studio podcast. In this edition, we have my friend and partner Ed Black. Ed, I’m so glad to be with you here today. You’d originally been slated to ride off into the sunset as a retiring partner at the end of 2023, but instead, you’re going to be leading a new initiative here at the firm that is dedicated to AI. I’d love to hear a little bit more about that role and what you’re going to be focused on.
Ed Black: Melissa, thank you so much for having me. It’s great to be here. I was slated to head for retirement. For the past—depending on how you count—between 30 and 35 years, I’ve been working as a lawyer with clients on how they carry out their technology goals and technology strategies in all their different businesses. As I was in my final year, I began to have conversations with the firm about how I might be able to continue to add value helping Ropes & Gray facilitate and develop its technology strategy. We’re in an environment now where, in part because of AI but also for lots of other reasons, technology is being aggressively integrated—not only into the back office of the law firm, where it’s been around forever, but into the front office of the law firm. It’s being integrated into how we serve clients and what clients expect to see, and that’s a place where I hope to add value in my new role. No longer as a client-facing service provider, my job title is technology strategy leader, and I’ll be working with people across the firm to make sure Ropes is at the forefront of how it uses technology in its front office services.
Melissa Bender: That’s great. I’m super excited about that, and we’re going to dig into talking a bit more about the strategy and the service offering later in our conversation. But first, I’d like to take a step back. I think it was a year ago that I first heard about ChatGPT. Since then, we’ve been hit with a tsunami of AI in the news, including in conversations with clients and likewise here at the firm in terms of engagement with AI. But AI isn’t new, right? What’s so different and new about the ChatGPT technology that this conversation has reached a new and different level?
Ed Black: Melissa, I’m completely with you in this idea that I think you’re suggesting, that ChatGPT is as important in the sense that it’s a cultural moment as it is in the sense that it might be a technology moment. AI technology, in particular the type of AI technology that’s involved in the open AI initiative ChatGPT—large language models and machine learning—that technology has actually been around not just for years, but for decades. It’s been advancing very incrementally, very consistently, but what ChatGPT unveiled was how far it has come over a period of years and how useful it could be not just to handle niche problems defined within large organizations, like the Defense Department or giant companies, but how useful it could be to ordinary people doing ordinary things that we all do every single day. Suddenly, a bunch of people in the legal profession and outside of the legal profession realized—in a way that maybe they could have realized sooner, but in any event ChatGPT underscored it—that many of the day-to-day tasks can now be carried out in a way that is augmented by if not replaced by artificial intelligence tools. And that was the cultural moment. There’s a whole world of people in and outside the legal profession who suddenly started looking at transitioning the performance of various types of tasks onto an augmented AI environment or fully replacing an AI environment, who up until that moment—up until they saw how ChatGPT could help a ninth-grader write a high school essay, they just weren’t thinking—they didn’t realize that the technology was there and had come that far.
That cultural moment of realizing how useful AI and, candidly, other technology tools are, opened up the door to a lot of questions that we get in the law firm now every day about how we’re going to use AI to be time and dollar efficient. Tasks that maybe took us 10 hours a year ago, why isn’t that taking us five hours now maybe with augmented technology? How we’re going to get to more comprehensive and more accurate results. One of the great strengths of technology, including AI, is the ability to marshal large data pools and deliver targeted insights quickly and efficiently. If you look at that question, “How to marshal large data and deliver a targeted insight”—that’s core to the legal function. People come to us all the time and say, “I’m buying a company.” “I’m selling a company.” “I’m forming an investment fund.” “I’m initiating a major piece of litigation and I want to know what the market is for this particular term. I want to know where the courts stand nationally, if not internationally, on this very specific issue. And I want to know it now and I want to know it at the lowest possible cost.” So, that’s how we’ve engaged at Ropes. We’re looking with a large team—and I’m facilitating but many others are involved—to try to target AI and related technology to the delivery of legal services in a fashion that maintains excellence, is compliant with our regulatory environment, ethics, malpractice, etc., but is really time and dollar efficient for the client. It’s faster, cheaper, with high quality results, of the type that people associate with a Ropes & Gray work product: premium results.
Melissa Bender: I just have to say, what you’re describing really underscores why it is that in the venture industry we’re seeing that if you’re running a legal tech startup that is promising to use AI, you’re in pretty good shape to get funding, maybe relative to a lot of other folks that are launching new tech businesses. If you were to take a crystal ball and be reflecting on some of the ways in which AI is going to be disrupting the legal services industry, I mean—you’ve already alluded to some of the ways in which it could change things—what are some of the areas that you think are really ripe for change based on the adoption of this technology, including by law firms.
Ed Black: First, I think there are certain tasks that are carried out based on widely available, publicly available information, where in the end—we aren’t there today, maybe a 12-to-24-month timeline—we are going to see replacing technology. The reason I want to hit upon that point about publicly available is—remember: the usefulness and the accuracy in particular of the type of AI that’s really captured the public attention right now, this generative AI out of large language models—it really depends on the availability of truly large data sets that are representative and can result then in the AI training accurately and delivering accurate results. There are lots of areas where, candidly, AI could do more but in the immediate horizon probably won’t do more, because marshaling the training database could be tricky. But there are certain legal tasks associated with certain types of documents that are very commonly used—and when they’re used, they’re often filed in the public record, like with the Securities and Exchange Commission or elsewhere, and where large data pools can be assembled. In that instance, you could see—in the 12-to-24-month timeline—migrating toward a world where humans are not gone, but there is an augmentation of the drafting process and the review process that allows you to very quickly assemble market-based thinking as to what the documents should look like as compared to other market exemplars. It allows you to review drafts that you’re sending or drafts that you’re receiving, compared not only against the immediately prior draft, but against this large database, and allows you to seek and find counterproposals, new language and amendments that can be put in.
Remember the so-called hallucination problem. AI is software—it follows instructions. The instructions it’s getting are about what is the most statistically probable answer or next word in that sentence or next paragraph on that page in light of all of the history I, the AI, see in my training database. When it hallucinates something and says something is true, something is there or something should be revised, all it’s really saying is it’s the most probable thing in light of what was in its training database. And so, you still need a human there to impose that judgment and to make all those corrections, and candidly, that’s probably not going away anytime soon. You hear people marketing that they have technology that prevents hallucinations, and they do have guardrails that really prevent obvious problems, but that nuanced judgment, that something is just not the right tone, that’s going to take humans for a long time to come.
Melissa Bender: Yes, that’s something we see on the asset management side. We have been piloting quite a number of software tools that are available from third-party vendors who Ropes & Gray might be looking to partner with. I think in many cases, what we see is that those tools can be extraordinarily helpful in terms of cutting down some of the raw labor associated with very representative, time-consuming tasks, but that effectively the software has to be set up correctly to be asking the right questions if you’re going to get a good output. And then additionally, there’s nothing that’s really going to be substituting for having that second set of eyes, scrutinizing and looking for these hallucinations or other anomalies that may come out.
Ed Black: You’re absolutely right. And this highlights one of the challenges that Ropes & Gray is already planning to address. When you think about one of the things that AI and other technologies are migrating into the legal front office, one of the ways they challenge the law firm is they challenge the law firm to be a great partner with technologies and technology service organizations, and that includes technology service organizations that are making tech-based products that then we can bundle up with our Ropes & Gray services and insights to offer clients, but also to be a great partner with the technology service organizations that are inside our clients. So, we’re looking at developing and extending collaboration tools in a fashion that will really create a seamless connection between Ropes & Gray as a service provider and our clients. That’s going to require us to become much savvier in partnering with third-party providers that are both used by us and by our clients to extend collaboration tools and other digital environmental features. And there’s a pricing challenge there. There’s a pure technology cybersecurity-type challenge there. There’s, as I mentioned, a compliance challenge there: ethical rules, malpractice guidance, our client instructions. There’s a challenge there about meeting our client expectations.
One of the things that we in the legal service industry are going to be challenged by right away is the difference between where we are technically and where the service organizations that our clients deal with every day, where they are. I have an app on my phone where I can immediately go look into my bank accounts—by the way, information I regard as critically private and needing to be protected by the strongest cybersecurity—and while out and about using my phone app, I can look at images of every check I’ve written and records of every transaction I’ve engaged with for the past three to five years on my phone. Now, all of our clients, of course, can do the same thing. I think one of the cultural moments that ChatGPT brought about is the realization that there are tools that will marshal these large data sets. We have clients who are now wondering, “Why aren’t my documents available at my law firm on the same basis?” And we’re going to rise to that challenge. We’re going to integrate with our clients in a way that makes it faster, cheaper, and frankly more convenient for them to take advantage of our legal services.
Melissa Bender: Yes, I couldn’t agree more. One of the things that I and others have been thinking about here at the firm, within asset management in particular, is what is the value proposition that we in Ropes & Gray are bringing to our clients from a legal tech perspective? Increasingly, what we see is that asset managers are themselves reliant both on external vendors and external service providers that are developing software solutions in many cases that may be based on AI or utilize some form of AI, but likewise they are increasingly investing in terms of their own internal infrastructure—not only to be wanting access to their data sets and the ability to sort and see things on Ropes & Gray’s system, but on their own systems as well, and how is it that they can more efficiently conduct all aspects of their business, including compliance, which is increasingly onerous. Other partners and I have really been focused on is understanding the technology tools that clients are utilizing in terms of external service providers, but also what they’re building internally, so that we can really be advising with them and consulting with them in partnership around those technologies.
Ed Black: That’s completely right. You talk about where is the value add. What is it that Ropes & Gray can and will continue to provide? One of the challenges that actually our clients face most acutely is that a lot of this new technology is very expensive. It’s expensive to set it up. It’s expensive to maintain it. It’s expensive to monitor it and keep it running in compliance with not just general applicable laws, but whatever special rules might apply in a given industry, this is a real opportunity for Ropes & Gray to help our clients. We do have some organizations that are so large that they can absorb the full capital cost of setting up their perfect digital environment just for themselves as though it were a private playground, but the fact of the matter is that the vast majority of our clients don’t have or wouldn’t want to allocate the resources necessary to build that perfect environment where the only users sharing the cost is just them. And we have an opportunity at Ropes & Gray to build a technical environment. Again, it’s going to challenge us around our ability to be a good partner for the service firms and a good partner for our clients. It’s going to challenge us, again, around pricing. It’s going to challenge us in our compliance environment. But we have an opportunity to be a focal point for our clients and create an environment that they can essentially cost share and community share through use of the platform that Ropes & Gray creates, allowing even some of our smaller clients to immediately be using these tools in a fashion that moves them right to the vanguard in terms of what’s available and how useful it is to them. We’re hoping that as we move forward into 2024, we’ll be able to actually see some of these things becoming implemented. It’s a long road. And again, lots of challenges. We’re aware of them. We’ve got a team. It’s definitely not a one-person effort.
Melissa Bender: As our clients start dipping their toe into the AI waters and thinking about the ways in which they may be able to leverage these technologies in their business—obviously they can come and talk to Ropes, we’re happy to be engaged with clients on that front—what other suggestions would you have for clients as they’re looking to really start engaging more actively with what the opportunities may be for their business here in terms of leveraging this technology?
Ed Black: The first thing you mentioned, I just want to repeat it because it’s a very important part of our message, is we are ourselves a member of the user community for this technology, and we are doing our best right now to be that good partner. Things that we already have going that our clients might want to jump in on are peer-to-peer roundtables. One of the things that I’ve found most useful as I try to move toward technology strategy is the ability to call people in other comparable organizations and say, “What do you think about this question? What are you doing about this question?” And so, we’ve set up for AI governance a series of peer-to-peer roundtables where we have groups of 10, 15, 20 people, all of whom have responsibility in their organization for figuring out how they’re going to overcome some of these challenges I’ve identified: the compliance challenges, the economic challenges and so on. We have these peer-to-peer discussions—Chatham House Rules, nothing leaves the room—and we’d be happy to add more clients to that dialogue.
Melissa Bender: These roundtable groups, are they organized by industry, geography? How are you thinking about tackling that?
Ed Black: The partner in charge is Rohan Massey, and he currently has the plan moving forward with both industry focus and geographic focus. One of the things we’ve found in these roundtables is that a roundtable of 1,000 people is useless. We really also do want it to be peer-to-peer discussion. If you’re dealing with how to manage AI in emergency room medicine, do you really want to talk to a bunch of peers who are using AI to run hedge funds? Interest has been so strong in these peer-to-peer roundtables that so far, we’ve seen no problem at all creating groups that people want to be in and that they are finding useful. Now, by the way, even without the roundtable, Ropes & Gray itself is happy to be part of that conversation. We did this when COVID came along—there was a world of regulation, policies and procedures that needed to be invented on the fly to shut down your office, to move everyone off site (I won’t repeat it all). In any event, we had to invent all of that for Ropes & Gray and became basically a globally recognized thought leader because we were so committed to the partnership element. We made all of these things available to everyone who participated in dialogue with us, and we’re doing the same thing with AI. Where we’re very happy to jump in just with our own policies and procedures, and again, we’re developing them. Many challenges still lay ahead for us, so it’s not a completely done deal. It’s not one of these things where we can send you the checklist—we’re working on a checklist. But happy to be in a dialogue.
The other things that we’ve got going is we have a group coordinated by a colleague of mine—not a lawyer—an IT professional, an engineer, a fellow named Sergey Polak, where we’re very actively monitoring and being in close touch with service providers providing AI-based products in the market. We have a number of clients who are in that field—we have a number who are both users and publishers, or producers and providers—and we stay in very close touch with the market. We’re happy to talk about what we’re seeing and how that’s going.
We also are thinking constantly, again, about the economic models, how the cost infrastructure for this works. Fortunately, we have a great team inside Ropes & Gray in our finance group, but we’ve got a couple people who are now full-time involved in technology strategy, figuring out those issues. We have a great knowledge manager, Patrick Diaz, who’s focused on the use of AI and knowledge management—he’s available for discussions. We have a CIO, Marsha Stein, similarly available. So, lots of points of contact depending on your angle. Again, we also hope to rise to the challenge ourselves and to be announcing in 2024 how we’re adopting policies, products, and services that really make the most of technology in our front office, in our client-facing services. That work product is something we’ll be looking to our forward road map for.
Melissa Bender: I assume that if any of our listeners are interested in learning more about you, this practice or any of this menu of connections that you’ve just offered up, folks can reach out to you?
Ed Black: Absolutely. Now, a couple of things, in fairness—I’m talking now about people who are thinking about the business and practical issues in consuming these technologies, services, AI-based services. But we’ve got people in our health care group who are advising, on the legal end of it, health care use of AI: Christine Moundas and Megan Baca; AI and data privacy: Ed McNicholas and Fran Faircloth.
Melissa Bender: Alright, well, this has been incredible. Maybe we’ll have to do a part two and we’ll do a check-in in six months and see how some of these things are going, because I would love to hear an update. But in the meantime, I would love to know whether you’re a new year’s resolutions kind of guy, whether there’s anything on your retirement bucket list that you’re still hoping to knock off in 2024. Anything that you’re looking forward to on a more personal level?
Ed Black: I’ve been a working lawyer at so-called big law for over 30 years, and so, back when I thought I was retiring, I booked a multi-week trip to the New Zealand Alps, to the mountain country. And so, when I signed on for the new job, which is a job I’m very excited about, by the way, but I said right from the get-go I’m not canceling, so I am going to take a few weeks to go to New Zealand and clear my head—that’s certainly on the agenda. And then, my new year’s resolution is actually tied to this new job I’ll be taking. The thing we want most out of the new initiatives, because we’re a service organization, is happy clients, and so, I plan to be traveling quite a bit to actually visit and meet with individual clients to make sure I understand their expectations of us in the area of technology and technology strategy. Hopefully some people will be kind enough to take me up on it and sit for an hour and make sure they let me know what they think.
Melissa Bender: That sounds fantastic, and you’re going to be racking up a lot of frequent flier miles in 2024.
Ed Black: Fortunately or unfortunately—yes, exactly.
Melissa Bender: Well, great. Thank you so much Ed and thank you again to all of our listeners. This has been the R&G Tech Studio podcast, available on the Ropes & Gray website, on the R&G Tech Studio podcast page, and also wherever you get your podcasts. Thank you all for listening.