Vanderbilt Law School recently launched an exciting new initiative called the Vanderbilt AI Legal Lab (VAILL) to explore how artificial intelligence can transform legal services and access to justice. In this episode, we spoke with VAILL’s leadership – Cat Moon,(👑) Director of Innovation at Vanderbilt’s Program on Law and Innovation (PoLI), and Mark Williams, Associate Director for Collections and Innovation at the Massey Law Library – about their vision for this pioneering lab.
VAILL’s mission is to harness AI to expand access to legal knowledge and services, with a particular focus on leveraging generative AI to improve legal service delivery. As Moon described, VAILL aims to experiment, collaborate widely, and build solutions to realize AI’s potential in the legal domain. The lab will leverage Vanderbilt’s cross-disciplinary strengths, drawing on experts in computer science, engineering, philosophy, and other fields to inform their ethically-grounded, human-centered approach.
VAILL is prioritizing partnerships across sectors – courts, law firms, legal aid organizations, alternative providers, and others – to test ideas and develop prototype AI applications that solve real legal needs. For instance, they plan to co-create solutions with Legal Aid of North Carolina’s Innovation Lab to expand access to justice. Moon explained that generative AI presents solutions for some legal challenges, so VAILL hopes to match developing technological capabilities with organizations’ needs.
Ethics are foundational to VAILL’s work. Students will learn both practical uses of AI in law practice as well as broader policy and social implications. As Williams emphasized, beyond core professional responsibility issues, VAILL aims to empower students to lead in shaping AI’s societal impacts through deeper engagement with questions around data, access, and algorithms. Teaching ethical, creative mindsets is VAILL’s ultimate opportunity.
VAILL will leverage the resources and expertise of Vanderbilt’s law librarians to critically assess new AI tools from their unique perspective. Williams noted that the lab sees law students as a “risk free” testing ground for innovations, while also equipping them with adaptable learning capabilities to keep pace with AI’s rapid evolution. Rather than viewing AI as a differentiator, VAILL’s goal is producing legally-skilled innovators ready to thrive amidst ongoing change.
Vanderbilt’s AI Legal Lab represents an exciting development in exploring AI’s legal impacts. By emphasizing human-centered, ethical approaches and collaborations, VAILL aims to pioneer solutions that expand access to legal knowledge and services for all. We look forward to seeing the innovative applications VAILL develops at the intersection of law and AI.
Listen on mobile platforms: Apple Podcasts | Spotify | YouTube
Contact Us:
Twitter: @gebauerm, or @glambert
Threads: @glambertpod or @gebauerm66
Email: geekinreviewpodcast@gmail.com
Music: Jerry David DeCicca
Transcript
Marlene Gebauer 0:07
Welcome to The Geek in Review. The podcast focused on innovative and creative ideas in the legal profession. I’m Marlene Gebauer.
Greg Lambert 0:14
And I’m Greg Lambert. So Marlene, we, we had to skip last week because I was recovering from a bout of COVID. I did the wonderful conference COVID But apparently I mean, I talked to a number of people, and I was the only one that seemed to have caught it. So.
Marlene Gebauer 0:31
Well, I got the conference cold. So I didn’t I didn’t get COVID though.
Greg Lambert 0:36
I’m feeling I probably caught that I got the flu. I cut COVID I caught you know, everything
Marlene Gebauer 0:41
got it all taken care of.
Greg Lambert 0:42
I’m super strong now. So
Marlene Gebauer 0:45
Nothing can harm you.
Greg Lambert 0:48
And we’ll see. We’ll see. So I understand you had some uncovered travels yourself.
Marlene Gebauer 0:54
Yeah, that’s true. I just got back from an advisory board meeting for the IOC for Wolters Kluwer. And that’s if anyone doesn’t know. I mean, that’s, that’s basically a group of innovation and KM professionals. And we work quite closely with the Wolters Kluwer folks about feedback on products and you know, where the industry is going. And so it’s always very a it’s always a very insightful meeting. And this was no exception.
Greg Lambert 1:26
Yeah, yeah. I like it. A lot of the vendors are, it used to be more, but there’s still a number of vendors that have these advisory boards that actually reach out to their customers and say, what’s working? What isn’t. Here, test this.
Marlene Gebauer 1:39
is really critical to sort of hear from your customers in terms of what’s working, what’s not what their clients are saying, you know, in order to make smart decisions. So
Greg Lambert 1:50
yeah. Well, maybe we should suggest to law firms that we have a client advisory board. What do you think?
Marlene Gebauer 1:57
Yeah. I think that’s every day. Like, we’re we’re on the advisory board every day trying to find out what they’re thinking.
Greg Lambert 2:06
Yeah. Well, this week, we have a fascinating conversation lined up for everyone with two of our favorite legal innovators, who are at the forefront of exploring the impact of artificial intelligence on legal services and access to justice. So we have from Vanderbilt Law School. We’re excited to welcome back our five now sixth time returning guest, Cat Moon, Director of Innovation at Vanderbilt’s poly, which is the program on law and innovation. Cat. Welcome back.
Cat Moon 2:39
Hey, y’all. It’s good to be here. And I’m sorry, I’m not wearing my tiara. I’m in my office. Sorry. I’m in my office. And my tiara is at home. And I failed to get it. I’m so sorry.
Marlene Gebauer 2:56
Everyone would be jealous at work if you wear it, so you can’t wear it at work.
Cat Moon 3:00
I know! Next time? Well, I promise. It’s beautiful. It’s beautiful.
Greg Lambert 3:07
Thank you.
Marlene Gebauer 3:09
Greg picked it out himself.
Greg Lambert 3:10
I did.
Cat Moon 3:12
Well done.
Greg Lambert 3:13
I did have to explain to my wife, why are you buying a tiara?
Cat Moon 3:20
That’s fair. That’s fair.
Marlene Gebauer 3:22
Well, we also want to welcome Mark Williams. Mark is the Associate Director for Collections and Innovation at the Massey Law Library. Mark, you’re only former interviews away from getting your own tiara, if so choose. So welcome to The Geek in Review.
Mark Williams 3:37
I really look forward to that. Hopefully we can acquire. In the meantime, I appreciate you giving me equal billing to Cat we all know I’m just kind of drafting off of her popularity here and I’m more than happy to do so. So
Greg Lambert 3:51
yeah. But I saw yesterday or sometime this week that King Charles was wearing his crown so I can now envision what it looks like Mark. Yep. Yeah.
Mark Williams 4:03
We could just do the smoking jacket for me. Like Saturday Night Live. There we go. We’re both really
Marlene Gebauer 4:12
okay to mix and match. Yeah. All right. So
Greg Lambert 4:15
well as if the two of you weren’t wearing enough hats head to Vanderbilt. You’ve recently launched the Vanderbilt AI Legal Lab, or VAILL. VAILL is in Vail, Colorado or piercing the corporate. So
Marlene Gebauer 4:30
Actually there were people I looked up on the internet that have the same spelling, what is it actor and one is an author. So just letting you know. So VAILL is an innovative innovation launched by Vanderbilt law school to explore and understand how artificial intelligence intersects with and can transform the delivery of legal services and access to justice. It’s housed within the schools program on law and innovation. Cat Would you mind telling us what the mission of VAILL is?
Cat Moon 5:00
Yes, so Well, I will preface this by saying that the adage building a plane while you’re flying, it certainly applies here. But I really think we’re like everyone else right now that the mission really is primarily to go forth and see what we can build. We are framing kind of a formal mission statement. And right now, it’s somewhere around this. We have our human centered mission to harness AI to expand access to legal services, and legal knowledge. And I think primarily to focus on how we can improve the delivery of legal services, through the use of AI. And for probably obvious reasons, were pretty focused on generative AI right now and exploring and seeing how that’s going to evolve. And I think we’re at this point in time, that there’s just so much potential and opportunity. And for us to really, truly harness and make the most of it, we just gotta move fast. And I will not say move fast and break things. Although we definitely are a scrappy startup. But, you know, the, really our entire mindset is how can we experiment and figure out how this can work, what the technology can do. So so much opportunity.
Greg Lambert 6:24
And Cat, I had the pleasure of going out a couple of months ago, and talking with the your class that the you and Mark teach on essentially generative AI, and now it’s playing a role in the education of the students as well as what the potential is for the market. And in fact, Mark, we were we were joking before we went on. So the three of us did this fantastic interview for the podcast. But somehow or another. We were doing it at a bar. And somehow or another, I did not press record.
Marlene Gebauer 7:01
Operative phrase: “We were doing it in a bar”.
Greg Lambert 7:06
They were unrelated. But it was a great conversation. So
Mark Williams 7:09
that was a great conversation the lost tapes.
Cat Moon 7:12
We sounded so smart.
Mark Williams 7:14
I’m not gonna be able to recapture that here.
Greg Lambert 7:19
Oh, well, that was one of those things where, you know, as a as a Gen X, or is like just experienced the moment just enjoy that it happened? Well, Mark, you considering your role there as the Associate Director for the Law Library? Do you mind just sharing a little bit about how VAILL plans to leverage the library resources and collections to foster the innovation for the program?
Mark Williams 7:48
Right, you? Well, I think, and you both would know this, it’s just, it’s just an extension of the work that law librarians have been doing forever. It’s sort of a natural, natural evolution and to be able to evaluate, and sort of critically use these resources, and wanting to make the law library a place where, you know, people can try stuff out and experience and have the resources to do so. At least did while you’re in law school. And it’s somewhat of a risk free environment, I forget who was one of the cofounders of CaseText. I can’t remember which one it was talking about, like if the best compliments he had received on CoCounsel was just that. The law librarians kind of said, well, it was okay. Like
Marlene Gebauer 8:35
we know exactly what you’re talking about. Yeah, that’s
Greg Lambert 8:37
A motto of my old, old home state of Oklahoma. It’s OK.
Mark Williams 8:45
Yeah. So I think, you know, that’s a lot, what we bring is our perspective, and also our resources to just be able to critically evaluate these tools. And the same way that we’ve been doing for ever, in this different framework, and be able to provide valuable feedback, and Cat and I actually met with a startup of another generative AI legal company today and talking about sharing their beta product with our law librarians and also with our students, and we just want to be a place where groups like that feel like they can come and try stuff out and get feedback from us. So I think that’s a huge opportunity for us. And for our students to be able to have that sandbox environment to just again, have an adopt a mindset of continuous learning. What questions to ask, what workflows work, what don’t, what models are you built on, you know, to critically evaluate the exhibit? Usually have not been very sexy, but law librarians have always cared about but I think more people are going to start caring about if that makes sense. I think that’s the approach we can bring.
Greg Lambert 9:50
Yeah, like, I like that a lot. I and I think that’s something that like, in the 90s when the when you know, when we’re getting into more Web based type of materials, that the law libraries at law schools were the proving grounds of a lot of these things. And it felt like like the last 15 years, that’s kind of fallen out a little bit. So I’m glad to see that that seems to be coming back with the generative AI tools.
Marlene Gebauer 10:21
Yeah, I mean, it sounds like a good preparation step. Because I can tell you, you know, in firms, people are clamoring for that type of education, you know, they they want, they really want more knowledge about how these things work, what they do, what things should they be concerned about? This is great for students that are kind of coming into the profession, to sort of be armed with that information already.
Mark Williams 10:44
Yeah, we really feel like there’s been almost an inelastic demand and curiosity for learning about this, both from our students and from just professionals that we’ve talked to. And that’s, you know, a big impetus for why the lab, we got the momentum to do it is because of that, for sure.
Marlene Gebauer 11:01
Cat, you’ve mentioned in recent press releases the importance of cross sector partnerships for extending the impact of VAILL. Can you elaborate on what kind of partnerships you’re aiming for to forge and how these collaborations might shape the lab’s work?
Cat Moon 11:16
Absolutely. And I’ve to preface my response with respect to the lab specifically, I do want to point out that it’s been the kind of tradition and in part, the reason for existence of the program on law and innovation, to figure out how we forge collaborations across when I call silos, I guess, four or five years ago, now we started the summit on law innovation SoLI. And that was particularly aimed at creating opportunities for folks across these silos to come together and kind of cross pollinate, right. Unfortunately, as with many things, COVID threw a wrench in that happening. But the lab, I think, is the next iteration with that. So the goal really is how can we bring people first of all across disciplines, so we have an embarrassment of riches, frankly, here at Vanderbilt, with respect to computer science and engineering, people doing really interesting work in the humanities with respect to AI. Actually, Mark and I sat in on a fascinating conversation yesterday afternoon, involving some folks from philosophy from history, just doing some really interesting things. Not to mention just the folks in our building who’ve been doing interesting things with respect to AI for a long time. The opportunity to bring people across Vanderbilt into the legal ecosystem. But extending beyond that. Our goal with the lab is to create collaborative opportunities, with courts, with law firms, with legal aid organizations, with smaller providers, hopefully with alternative legal service providers. So we really are in this active sort of listening phase to understand who’s interested in experimenting and collaborating on some projects. I can tell you, we have already in the works, a collaboration with Legal Aid of North Carolina’s Innovation Lab, they’re the first legal aid organization to have a dedicated lab with a director, a full staff with some really lofty goals. And so we’re going to be collaborating with them. We are talking to a number of law firms who are doing some interesting things to find kind of the right mix of opportunities to experiment and collaborate. They’re same thing with some courts. And so we’re just actively seeking out people who are curious and interested in trying some things and building some things. And so my hope is, it’s going to be a really compelling combination of how we can harness and bring together resources across the developed and then take those out into collaborative opportunities across the legal spectrum. We just want to delve some stuff. We want to do. Right? Like Enough talking, we just want to do.
Yeah, that makes sense. I’m just curious. When either you reach out or these organizations reach out to you whether they’re alternative legal service providers, law firms, or legal aid, whoever it is, how do you pitch to them? Here is the win for both of us. Here’s, here’s what you can bring. Here’s what we can bring. How do you pitch that? I know it’s new. I know it’s new. But if I were to come to you from Jackson Walker and say Cat, I really want to, I want to leverage what we’re doing with what you’re doing. How do you sell me on that?
I think ours it’s by listening and understanding what you all would be interested in doing and And then, I mean, we I don’t want to speak for Mark. So I’ll let him chime in on this as well. But one thing we’re actively focused on right now is creating that Nature Explore what is it that we have to offer, institutionally and process wise through the lab, and then kind of filtering through that the collaborative opportunities that come our way? Right. So as I said, we are building the plane while flying it. And in part of it is also just really getting a handle on what other folks are doing and where they are in their processes. So I’m coming at this, again, speaking only for myself, but I’m coming at this with a very much of a growth mindset. Like I know enough, I’m I’m a humanist who dabbles in technology. That’s a phrase that I definitely latched on to from this interesting conversation that Mark and I had with some other folks here on campus yesterday, with respect to AI, specifically. So I feel like every conversation we have, I learned something more about what the opportunities look like and what they are. And so I guess, because I always have to quote Mary Oliver, I’m definitely leaning into living in or being an open mindedness of not knowing enough about anything, right. And so that That certainly applies with respect to that, but so that we can have some productive collaborations, we are working towards having this this matrix that can help us really identify. And so then it’s a matter of in terms of pitching, it’s a matter of when we can fully understand what we bring to the table and understand as much as possible what the goals of another organization are, then it’s like just looking at what does that Venn diagram look like? Right? Where’s the intersection? And does it make sense to try to do something? And I will stop rambling now and see if Mark has anything you’d like to add?
Mark Williams 17:07
Yeah, especially with Cat likes her focus for years on on human centered design, and that process like that work continues. With this, this is perfectly suited to that, because I think, right now, like a lot of the demand we’re seeing, the collaborations are things we have to flesh out a little bit. It’s exactly what the nature designing those problems is gonna be what the use cases are, and that human centered design processes like central to that. But there’s also just right now such an insatiable curiosity for what the tools are and what the use cases are. That just that education component right now has been so important. And it’s part of it. I think we talked about a lot too, as it’s just we’re still just defining what the questions are. Because if we’ve soon as you think you have the answers, like even you just saw OpenAI’s developer day, Monday, like you think you have the answers, and then the questions change. So it’s, it just becomes just an ongoing sort of learning environment in that way. But yeah, I think that’s direction, we’re going into sort of equipping our students to help us out with that process as much as we can.
Marlene Gebauer 18:15
That’s a really interesting point, because again, sort of building the airplane while flying. I mean, I know the mission includes, for VAILL includes training of students to sort of navigate the landscape. And so again, the coursework is involved. And so there is a degree of of prediction, I guess, in terms of what’s going to be important and useful. And I’m really interested in this because, you know, we are struggling, you know, on the firm side in terms of how much education what sort of education on Gen AI, given people are not, you know, most of the people that are attending this are not technologists, but yet they need to know, a certain level of knowledge about that. So, you know, Mark, can you give us a glimpse into the kind of coursework and hands on experience that students can expect from VAILL? And, how do you think that’ll prepare them?
Mark Williams 19:06
So I think one of the things that we noticed, and that allowed us to get off to a quick start is that we had already kind of had a suite of courses at Vanderbilt, that fit into what we could have called a concentration, even before we got started. So we have some additive things that we want to expand on, but like a lot of what we had already offered and we’re doing it at school was already here. And it was just a matter of like being intentional about it. So Greg came to our AI and law practice course. And really like every week of that course, every unit like eventually could be its own, like semester law course. I feel like that was that sort of wide ranging. So that was kind of a 100-level survey course. But the we have the ability right now just to run the gamut of having a very sort of What I call a I know how to drive the car pathway where they have star students can have an understanding of very basic stuff like just best prompting techniques, strategies using, the persona-shot prompt, few-shot prompting, like very nuts and bolts of this moment how best to leverage it Large Language Models. But then also understanding the underpinnings of the technology to avoid Avianca Schwartz-brief situation and how easily avoidable that was with just a little bit of technology literacy. But then also, I think what Cat and I are big on too, and that her other courses through PoLI that we can sort of already offer and think about through an AI lens is if AI is going to be a part of our world and is automating some of these low level tasks that used to take up our work. What then do we do with our time as attorneys? What are the high value things that we do with our attorneys that now AI allows us to do? Because we’re experts at AI, and we’re experts at law? How can we sort of use that to augment our practice and basically sort of turn that into an exoskeleton for your brain to basically have a superpower? So we’re looking at it from both that sort of nuts and bolts of this moment, version. And then the other courses that we I think we already have in place and will continue to add will be kind of a growth mindset? How do I continue to learn how to learn about this to leverage my already sort of high level of legal expertise. Another pathway that I’m hoping it’s a little bit further off into the horizon for something I hope we can accomplish is, because of the collaborations, we have Vanderbilt with computer science, to look into some interdisciplinary, sort of what I would call the build the car pathway, where for people who have that legal perspective and background, but also may have an interest in training models, and diving deeper into, you know, the ethics of data and algorithmic bias for more of a technical perspective, that is more far reaching. Now that is more aspirational than than realistic at this current moment. But we certainly have those sorts of practical underpinnings into the policy oriented, we already have a lot of professors doing a lot of policy oriented work here too. So you can learn the basics of profiting but then also sort of learn the Law of AI, I think is part of this concentration is how we envision it.
Marlene Gebauer 22:29
One of the things that I hear that keep coming up in discussion is training of the associates, like if like a lot of the some, I don’t know, this this type of work that that AI might solve for us is to an extent things that people have interpreted that sort of you learn through doing on some of these things. And if they’re not doing it, how are they going to be learning? So I’m curious if if that is something that is a discussion that comes up?
Mark Williams 23:00
I think that is a central question of the lab that we will I don’t necessarily have an answer for. But this certainly is what the driving force of creating the whole lab is about. And I think just looking to history can sometimes it’s can’t be completely, you know, this is still uncharted territory. But I think it’s Ethan Moloch, who’s a professor at Wharton, who does a lot of writing about sort of, he calls himself an AI pragmatist is very much how I view myself to likening a lot of this to sort of the introduction of calculators, in classrooms back in the 80s. It’s not a one to one comparison. But there again was like a reevaluation of what are the entry points? At what level? Do you need to learn these skills? Do you need to like struggle with the blank page to learn how to write a draft versus when do you bring it in? When are you reach the proficiency level? Or is this a writing task? What is more ripe for automation that isn’t worthy of this struggle of the of the blank page? I think that is a definitely a core mission. I wish I had more answers for you. But if I had the answers for you, then we may not need a lab.
Marlene Gebauer 24:08
I don’t think anybody has a definitive answer on this yet. It’s just something we’re all kind of wrestling with, like when to, you know, allow, you know, associates to use it like is there a certain level that they start? You know, there’s just a lot of questions.
Mark Williams 24:23
So much so much of what we train our associates on is potentially what this does very well. So I do think we have to find out what is our runway for achieving that proficiency without wasting anybody’s time?
Cat Moon 24:37
And if I can just chime in really quickly. This presents an opportunity that’s been right for a long time for our legal education and firms to think a little bit more intentionally about maybe how we’re working together to solve for this. So one thing I do in all my courses and it’s definitely imbued in the work we’re doing in the lab is empowering our students to really own their lifelong learning trajectory. So, in many ways, I think it’s a mistake to leave that up to a firm. And so that’s one thing I want to empower my students to do is to go forth and understand that, ultimately, it’s on them to figure out how they’re going to get what they need to be able to be excellent in their work. And there’s so much opportunity, I think, for there to be a closer collaboration so that law schools are thinking about what they need to do differently, and modify and tweak and maybe introduce. It’s and not or, and what firms need, can feed back into that and what that loop looks like. So I would identify that as a clear project opportunity for the lab, right to work with firms to help understand what can we be doing at the legal education level that’s going to prepare associates better to really thrive and succeed once they get into the firm environment? And likewise, what do firms need to be prepared to do? Just an observation anecdotal. That, you know, there’s always work to be done to improve how associates are trained? And maybe this was the runaway there. And the reason for that to be approached a little bit differently?
Marlene Gebauer 26:19
Well, I mean, I know one area, that that would be great for collaboration, I mean, is sort of the security and client related data and privacy and confidentiality issues, because again, that I think that’s another area where there seems to be a lot of question as to how how that works. You know, what are the protections in place? How do they work in terms of people because they’re, you know, there’s some hesitancy it’s like, am I allowed to do this? Am I not allowed to do that? So I think that would be an excellent opportunity.
Greg Lambert 26:52
Well, I’m going to either deviate from the flight path or shift down in our car that we’re building when one of those analogies here. So, Cat, as the Director of Innovation design, at PoLI, you are definitely no stranger to designing initiatives that kind of push the boundaries of legal education in practice. And you’d mentioned that one of the, and I think the reason for this is that this is finally a technology tool that can actually handle language. So you’re seeing more of the humanities involved in it. So I want to ask about, you know, how do you envision the ethical applications of AI that’s being pioneered there at VAILL? And what role will ethics play in the labs initiatives? How, how ingrained is ethics into the policies that you’re teaching?
Cat Moon 27:49
Like, I think it has to be a core feature, right, really foundational for a couple of reasons. So and to kind of center this on on the premise that VAILL’s work is in the AI of Law and not the Law of AI. So there is a distinction there to be made. And in a lot of ways, that distinction isn’t a realistic one, because it all kind of tends to sort of get mushed up together. But with respect to the ethics aspect, to me, that is a core piece of everything we do with respect to the course that Mark and I are teaching AI in law practice. And definitely with respect to the work we will embark upon through the lab. My hope is that we can think about it from an ethical perspective in myriad ways. So there’s obviously if we’re focused on the AI of Law, then we have to think about the ethical implications and just delivering legal services, right, and, and what all that entails. And you know, we have some amazing professional responsibility faculty who are super interested in looking into these things with us. So again, it kind of drawing on the embarrassment of riches that we have there at Vanderbilt, in the law school. And I think expanding that beyond, and this kind of goes into the law of AI. But again, I think making these distinctions at some point just becomes futile. Going from that into really thinking much more deeply with respect to cultural and social implications. I think ultimately, our goal is to prepare and empower our students to leverage these tools in an ethical way when they become practicing lawyers, but as maybe even more importantly, to have as deep understanding as possible, so that they can actually help lead and shape how we as a society are going to make choices With respect to how this technology is implemented. Really the stakes are much higher than how a law firm is going to leverage a technology platform. Right? When we’re talking about training lawyers. And so I think the the opportunity is enormous. But I think ethics is just an integral part of every single aspect of this. And to refer to a point that Mark made earlier, like we have a whole lot more questions than we have answers. Because I think as we watch the technology evolve, and it do things that even the people who are making it don’t understand. Yeah, it’s a lot. It’s a lot. But but the ethics piece is foundational, right? It’s like, it’s got to be a core piece, really, of every part of, of this work.
Mark Williams 30:50
And I’ll just add on to that. I think even just right now, Cat alluded to some of our our faculty already kind of working in this area. Cara Suvall, who runs our youth opportunity clinic here, but also does a lot of professional responsibility training, I actually did co presentation with her for a legal aid group couple months ago. And so we were already thinking deeply about, you know, the nuts and bolts of professional responsibility from ethical features of consent, confidentiality, competence, and how to avoid sort of, you know, the ChatGPT brief situation. But then, also for me, like I really nerd out about sort of the ethical implications of data, access to data, access to legal information, who gets to leverage that legal information to make products. Damien RIehl is the one who’s I think talked about this a lot, where it’s just a guesstimate of what it would take like $2 billion to extract everything, you needed two out of PACER to get the entire dataset in the implications for like, who gets to build legal products? And who gets access to that information? Based off of that structure? That’s a whole other ethical box that’s like different from the sort of the nuts and bolts Professional Responsibility one. So yeah, I think we are also like, still figuring out what are the ethics of these tools? When we don’t even necessarily know where we’re headed with them from week to week?
Yeah like, whether they’re open or closed models? I mean, there’s a lot of there’s a lot of question. So Cat, VAILL has plans to co develop service delivery solutions with legal aid organizations, as well as law firms and legal departments. Could you share some insights into how those collaborations are expected to work and what impact you hope they’ll have on access to justice?
Cat Moon 32:40
We are launching through my legal problem solving course, which is my course in human centered design that I’ve taught here since 2017, that we’re going to launch at least three projects next semester, starting in January. So one of those projects is going to be with Legal Aid of North Carolina’s innovation labs. So that’s the Legal Aid component. Another set of projects is going to be with the opposite General Counsel, here at Vanderbilt. So we are working with GCs, as well as their Director of Legal Operations, Lizzie Shilliam to identify some discrete projects that we’re going to work on with them. And then we are in conversation with respect to what that law firm piece is going to be. So we aren’t quite ready to announce that piece yet. I think the objective with respect to kind of this full spectrum of those projects is to employ the human centered design method, and certainly the mindsets to identify opportunities, and then start designing and prototyping solutions, right. Our focus is going to be on specifically leveraging generative AI with respect to those solutions. And that’s a little bit different than I will say that I normally approach in legal problem solving, because the objective always is to really get deep in the problem and the opportunity first, so you’re not working backwards from a specific solution you’re working forward from, do we really understand what the problem is, what the challenge is, and how we can best address and solve that. However, in this instance, I believe that generative AI in the direction it’s going because of now, its capacity with respect to words, and how that is what we do in law. Actually is constituting, it is a solution to some things, right? Like there are things that can do and is getting better at that will solve some problems for us. So in some ways, we’re just trying to kind of meet in the middle we have these challenges. We have technology that can do certain things has capacity. Capacity is improving with respect to other things that can do so how Can we match those things up? And you know, our objective really, is to try to build some things, what exactly that’s going to look like is going to depend on a whole lot of factors. We have resources we can leverage here, through capacity we have with budget and also resources. As I said, just embarrassment of riches. On Vanderbilt’s campus, we’re working with groups, who already are employing some resources as well. So ultimately, what that looks like, but our hope is, we’re going to be able to point to a concrete thing that we have prototypes are our prototyping, and say, here’s a potential use case, here’s a potential solution that is going to help solve a problem for this organization. And for the people in it.
Mark Williams 35:49
I think the fun thing about that process, too, is just what I all my favorite use cases with generative AI are the weird ones that I didn’t expect. It’s thought, the right meow, legal research memo or find this case for me. I mean, heaven forbid, but it’s all these sort of gray area and nooks and crannies, creative processes that you really have to tease out by talking to people, and what their workflows and needs are. Before you can really know. It’s the creative side, the weirdness of some of this that really impresses me the most.
Marlene Gebauer 36:26
So Mark, Cat, we’ve come to the part of the podcast where we ask the Crystal Ball question, and Cat knows the drill very well. Mark, this is your first time so I’ll let you go first. What challenges or changes do you see for VAILL taking on over the next few years?
Mark Williams 36:46
I think the challenge just as simply as as far as I know, we’re the first ones to try something exactly like this. There is no roadmap. But that’s the fun of it is there’s it’s a lot of adventure. Again, like one of the things we did in our class. I only did it when I needed to. But I pretty much every week I would do is like, well, here’s what’s new this week that didn’t exist the week before. And I feel like I didn’t set out to do it every week, but pretty much every week I did. And I think that’s gonna be as far as I could tell, gotta be continued with the way that we do this going forward. So I mean, if you’d have told me this time last year that I was sitting here, talking to you about launching an AI lab would have said, you know, you are the one hallucinating, not the lab, not the Large Language Models. So I am not in the prediction game right now. I’ve heard apologies for kind of reject the premise. But
Marlene Gebauer 37:40
We did say you could answer the question the way you want it.
Mark Williams 37:44
So I know, enough to know, I should shouldn’t be predicting too much about where this might be headed.
Marlene Gebauer 37:50
Cat, do you have anything to add?
Cat Moon 37:51
Yeah, I think I simply have absolutely no idea, Marlene, I have no idea. I do believe with every fiber of my being. And it’s really the only reason why I’m still like walking this path after 25 years in the practice of law, that we have an enormous opportunity to really shift how we think about doing our work, and really embracing some new mindsets that will serve us well going into the future. Well, I will say this, I’ve seen a series of articles that ask the question, you know, will will AI continue to be a differentiator? And I think that that is completely the wrong approach. That is like a second industrial revolution question to ask. What is a differentiator is the mindset that allows you to be flexible and agile, and change and adapt to change and deal with ambiguity. And so while I do believe this particular technology poses some unique opportunities in law, we didn’t really talk about the Access to Justice thing. So I’m just going to have a brief plug. I think there’s a really interesting opportunity there to create access to legal information to people who otherwise will never get that from a lawyer. But putting that to the side. The real opportunity here is to teach law students and lawyers, how to look at things a little bit differently, and how that can serve them in all ways, and not just in dealing with generative AI, right. That’s the real opportunity. The lab will succeed if we are able to teach our students how to successfully experiment and make law better. And if that can have ripples that goes out into the greater legal ecosystem. And whether that happens with generative AI or something else, we don’t know. This is just kind of that the current experiment we’re going to run so otherwise, I have no idea what’s gonna happen.
Mark Williams 40:02
What what Cat said, as usual? I’m just drafting off of her.
Greg Lambert 40:06
We all are.
Mark Williams 40:07
Yeah, I think, regardless of the technology, if I do have a crystal ball prediction, it’s that it’s that our students will come out of there with that mindset, which is more valuable than any technical training that of the moment that they can hit.
Marlene Gebauer 40:22
Agree.
Greg Lambert 40:23
Well, my idea to pitch to both of you is that somehow or another you create a Master’s of Law program on this so that you can have an LLM on LLMs. So, there we go.
Mark Williams 40:36
I’ve thought about it. We may have even talked about it in my office.
Cat Moon 40:44
That is the golden opportunity, Greg. Yeah, that right there.
Greg Lambert 40:47
Absolutely. Cat Moon and Mark Williams of the brand new Vanderbilt AI Legal Lab. Congrats on the launch of VAILL.
Mark Williams 40:58
Thank you.
Cat Moon 40:59
Thank you.
Greg Lambert 41:00
If anyone wants to learn more about it, where’s the best place for them to go?
Cat Moon 41:04
Well, we don’t have a website yet. We haven’t built that part of the plane yet. But it’ll be up soon. So I can’t point you there. That folks are welcome to connect with us on LinkedIn or drop us a line directly via email. We’ll share that info with you, Greg, Marlene, so y’all can post that in the show notes if you like. Yeah, just reach out.
Mark Williams 41:26
Yep.
Greg Lambert 41:26
That was good. All right. Well, thanks. Thanks to both of you for being on the show.
Mark Williams 41:30
Thank you very much
Cat Moon 41:30
Thanks for having us.
Mark Williams 41:32
Thank you for having us. Yes, absolutely.
Marlene Gebauer 41:34
And thanks to all of you listeners and subscribers for taking the time to listen to The Geek in Review podcast. If you enjoy the show, share it with colleagues, we’d love to hear from you. So reach out to us on social media. I can be found primarily on LinkedIn. You can also reached out to me on X at @gebauerm and on Threads at @mgebauer66
Greg Lambert 41:53
I can be reached at LinkedIn, on X @glambert and glambertpod on Threads. Mark and Cat to it other than LinkedIn, I’ll make sure that we just put your your LinkedIn posts or links out there is that okay,
Cat Moon 42:09
I’m @InspiredCat on Twitter.
Greg Lambert 42:11
That’s right.
Marlene Gebauer 42:12
I saw what you did there.
Mark Williams 42:15
I’ve been coming out of my shell more on social media. I’m used to usually a lurker. But yeah, let’s stick to LinkedIn. And I think you’ve just Mark Williams on LinkedIn.
Marlene Gebauer 42:24
And of course, the music you hear is from Jerry David DeCicca. Thank you very much, Jerry.
Greg Lambert 42:29
Thanks, Jerry. All right. Thanks, everybody.
Marlene Gebauer 42:31
Thanks. Bye bye.