Building Legal as the Company's Air Traffic Control: Ekumene Lysonge, NerdWallet CLO

Cecilia Ziniti01:04
welcome to the show.

Ekumene Lysonge01:05
thank you for having me. I really appreciate it.

Cecilia Ziniti01:07
All right, before we dive in, let's get in on this structure. So you combine you lead a combined trust and safety org. a little bit ago, I had Danielle Sheer, who created the title Chief Trust Officer at Commvault, and basically made the point that you have to move past legal to really have that influence. Now, you know, I I think this is a podcast about bringing great general counsel and being a great CLO, but In that structure that you created, was that part of it? This idea of being more than legal? And how has that shown up in practice?

Ekumene Lysonge01:44
Yeah, I I mean, you know, I I tend to talk to my team in terms of analogy. So so I equate the the legal kind of security function to being like air traffic control, if you will, the command center of the company. And so we're responsible for making sure crashes w we're avoidant of of crashes, making sure people are able to take off and people I equate to ideas, right? business strategies and initiatives are able to to take off kind of undisrupted and ultimately land those ideas squarely where they need to be landed in order to to deliver on whatever the value prom promise is to the customer. And so what I tried to do when when kind of I wanted to reimagine, because I had two silo teams, right? And most corporate environments. You know, the security team resides kind of in its own silo and and works primarily with engineering and and they're pretty unseen f in in most organizations, at least that's my my experience. whereas legal is similarly siloed, and then you have the compliance and privacy components thereof, and they also have their respective silos. And what I wanted to do is to make everyone feel as one team. And so I had to rebrand us under one umbrella to make everyone feel like at least for our own operating system, we were a part of one combined unit, and therefore we removed those silos or strength swim lanes that people had kind of predefined and we figured out how to weigh way better ways to work kind of seamlessly amongst ourselves and then present a united front to the company as we try to kind of push forward kind of this discipline that we wanted to push from a security compliance legal privacy perspective to the rest of the company. And so I rebranded the team trust and safety because on the legal side of the house, based upon how we go to market. I very much think that all the component parts of legal and that's inclusive of compliance and privacy and and legal operations are really there to preserve or ensure that the trust that our customers have have have generated in NerdWallet, they're they're there to make sure that we continue to protect that trust and honor it with the integrity that the company was built on over all these years. And on the other side of the house, I I saw the security team is the safety component. They're there to make sure that one, our our infrastructure and and our network and our our you know customer facing experiences and websites are secure and safe for our customers, as you alluded to you know, d we we serve twenty five million plus customers a month and therefore it we take that very seriously and want to make sure that every time a customer engages with Nerdwallet, they they do that with a level of of security and and safety about their experiences and the products and and all the ways in which they interact with our business. And so I brought the two together and and coined the term trust and safety and here we are. So

Cecilia Ziniti04:54
How how's that worked out? So I guess curious around like so a little bit about the nerdwallet, tell us a little bit about the nerdwallet business. But as I see it, I'm literally on the I'm on the website and it says 10 million financial decisions made, 15,000 financial resources, hundreds of experts, Chase, Sofi, American Express. So Nerdwallet, I mean, I love

Ekumene Lysonge05:01
Yes. Yes.

Cecilia Ziniti05:15
Great naming, but it's basically like someone at some point was like, I'm going to research the heck out of financial stuff. Let me put it on the internet and help others, right? So presumably that's how the company started. And then now, like, you're in some pretty specific areas, right? I'm deciding what credit card or mortgage or huge decisions to make, and I'm coming to the site. So with that background, and maybe any other background about the business.

Ekumene Lysonge05:25
Yes. Yes.

Cecilia Ziniti05:40
How has that trust and safety approach like affected both the business and in this case, the end users?

Ekumene Lysonge05:47
Yeah, so so at the heart of it all, NerdWallet exists to help customers make decisions with confidence and and and clarity about all a host of different financial subjects that they are confronted with in their life, right? And and and you alluded to some of those. What credit card should I get? What what who's the best lender for my particular mortgage situation? Where should I or how should I allocate my investment assets? And what's the best way to prepare for retirement? You name it, insurance products. And historically, Nerdwild was known as a place to go and learn about all different types of financial topics. We've evolved over the course of time to producing, you know, market comparison tools, calculators, and other things to help people evaluate that. But still very much in their own ecosystem, their own personal ecosystems. And then we ev have evolved even more recent history even further in the sense that we are doing it for customers now through special concierge offerings where we can actually route a customer to a mortgage expert, through a mortgage business that we acquired some time ago, or route a customer to an investment advisor to to actually serve them and manage their assets for them, or route a customer to an insurance expert who can help them kind of tailor an insurance offering for their particular circumstance. So we have, you know, the do-it-yourself option, we have the do it with you option with our comparison marketplaces. And then we have a do it for you option in a nutshell. And in it spans everything from credit cards to to insurance products and and everything in between as we we wanna get the customer more comfortable with turning to the nerds, if you will, for any and every decision that they have to make.

Cecilia Ziniti07:48
I love that turning to the nerds. So, okay, so is there a situation where I I mean, I guess this type of thing, you know, it it kind of gets to, you know, as a G Z, you want to be that trusted advisor. And so here your actual business, you're literally like you want people to turn to you to decide what mortgage is best.

Ekumene Lysonge08:02
Yes.

Cecilia Ziniti08:10
So you you can't really have security issues because then you're like, okay, you know, I don't trust this person with their own website, let alone to tell me what to do. And so how, you know, do you kind of staff accordingly in security? You know, there's been a lot of you know, there's a lot of different paths you can take, but where do you where do you turn that dial? And then obviously you're a public company as well. So tell us how you kind of set the risk priorities, I guess, at the company.

Ekumene Lysonge08:38
Yeah. so so there they're very they're varying types of risk kind of paradigms or or models that we we have to consider. Obviously, we have general frameworks like like NISTs that we evaluate our security protocols and our privacy protocols against. and and those are more for a board risk risk management, if you will, and risk observation or oversight. but then we obviously have all the the nuanced kind of ways in which to pres present secure environments or enable AI kind of assisted kind of environments to meet the consumer in the way that that our peers and competitors are meeting consumers, right? And so, you know, there's the kind of Oversight board oversight lens that we have to operate in the paradigm there that we have to operate within. But then we have to also operate within the confines of what is an ever-evolving, you know, a momentum-filled kind of AI-enabled world where customers are discovering products, engaging with products, and thinking about their financial circumstances in very dynamic ways, leveraging all the various LLM platforms out there. And so we have to still be able to meet them where they are and do that in a secure environment as well, which as I'm sure we'll get into, has presented a host of different challenges. All as I I like to ta tell my team, all challenges that you know, have a certain level of job security involved with them, right? Because they're going to, they're going to continue to evolve and they're going to continue to be present. And therefore you need prof professionals with experience and judgment to to help lead the c customer experience and also to to help lead the company into kind of that new new territory. So different kind of risk vectors, different risk meters, if you will, are are kind of ever present in in everything we do. And and I tend to be calibrated more from a a a risk tolerant, a high a higher risk tolerance level than most of the CLOs that I've in interacted with. And and I think because I come to most of conversations with a principled approach or a a perspective that is business centric first versus risk centric. And and that's just based upon my experience of having been actually an operator in different parts of my career. So

Cecilia Ziniti11:24
Any good examples of you mentioned AI, but you know, so you mentioned being a little little bit more I guess pragmatic about risk than maybe some of your peers. And, you know, in in the fintech space, there was a tweet that, you know, tweets sometimes you know they're jokes or whatever, but they have some truth to them that basically said something to the effect that like Robin Hood, you know.

Ekumene Lysonge11:33
Yes.

Cecilia Ziniti11:48
made a nicer UI on Fidelity and it's worth however many billion dollars. And it's like, you know, it's not, and I th I wanna say initially maybe Robin Hood didn't even have like the banking charter or whatever whatever it was. Like it what it literally was just a UI. And in the in Silicon Valley, people are like, wow, that's wonderful W UI, et cetera. But it what it what it what the reason I bring up that example is that, you know,

Ekumene Lysonge12:01
Right. Right.

Cecilia Ziniti12:12
just that slight frame or you can imagine that early lawyers at Robinhood had to be like, okay, yeah, we're gonna let people swipe left and right to make trades instead of, you know, clicking 97 things. And you know, now you get into Robinhood, and I will say they do a pretty good job where they're doing crypto, they're doing unusual things, and the warnings kind of look gamified. They got little, you know, gold pots of gold on them and things like that. And so I I I guess like

Ekumene Lysonge12:21
Yeah. Yes. Yes, yes.

Cecilia Ziniti12:38
Is that kind of the decision where your pragmatism shows up most? Or like like, is there something where you're sitting in a room and an idea is presented to you or your or your team where most lawyers who did not have the frame you described would be like, yeah, no, like just literally just have a knee jerk, not you know, that's a cool reaction. And so any any good examples, any any wins that you that you can share? And you know, obviously there's privilege and whatever, but if there's anything that you can share where you're like, yeah, most most people would not.

Ekumene Lysonge12:57
Yes. Yeah.

Cecilia Ziniti13:07
Agree to whatever it is.

Ekumene Lysonge13:10
I mean you can look at at our our website and a host of different experiences. We're a very test heavy kind of culture. We like to try things, they're always consumer centric in in terms of trying to advance the experience for the consumer and present the best product the best possible experience, the best possible product to the consumer w whet no matter what the vertical, right? And so I sit in rooms all the time and i I've I've for whatever reason, you know. When I entered kind of corporate America, always entered it from the perspective of companies don't grow without taking risk. And if you want to be in that environment, you have to be someone that enables it. It but I've always looked at it through the perspective of you need to be in a position to always say yes, but the yes has to be durable. It can't be something that's kind of a flash in the pan, that that doesn't stand up to scrutiny. and I think that comes with it. Or what comes with that is making sure that every time a decision is made, there's a certain amount of rigor and discipline around it. And you're just not always making decisions on the fly without it being principled and and based in kind of yeah, first principles, right? or general decisioning kind of frameworks that the company is as architected over time. And so yes, I I I am more of a yes first, but but but

Cecilia Ziniti14:23
Yeah, yeah.

Ekumene Lysonge14:35
Let's make it durable kind of person versus a no. And and I think that is in many respects kind of polarly opposite from peers of mine that I've

Cecilia Ziniti14:45
Interesting. So I mean I think it's it's also like this, you know, it's a little bit the one way door thing with Jeff Bezos too, of like, Okay, so if you're gonna go through a two way door and when you wanna come back, that's fine. But if you go through that two way door and you want the business to explode, what's it gonna look like? Yeah, that's great. I love that.

Ekumene Lysonge15:01
Yeah. And and in some cases, as I've told peers as we make decisions, certain things don't have a return. Like there is no return ticket. Once we go through this door, we need to go as one of the phrases I always use with with peers and in executive conversations is eyes wide open. And once something's in the decision box, certain decisions can't be re remade. and so it's the eyes wide open moment, guys. We're gonna do this. But just know that this is a one-way ticket to whatever the destination is and we all have to live with it. And we have to also make that representation to the board. Like directors, we are making a w a decision as an executive team that we believe has whatever exponential growth or w whatever the the end result is that we're we're pitching to the board as strategically aligned with where we want to go. we are making this decision as a group, and we understand that once we make this decision, it comes with some consequences that cannot be un undone or or rewritten, if you will. And and as long as everyone's locking arms and doing that together with their eyes wide open, I I don't have any issues with taking risks.

Cecilia Ziniti16:08
Similarly, what so one of there's a lexology article called Discipline is the Differentiator, a GC's Guide to AI governance. And so bringing this into kind of the AI world, what you're describing with the eyes wide open and the kind of like making sure that that a risk taken is is taken because you're aware of it, not because you didn't you didn't know. how has AI either changed or amplified? Or the ability to drive that discipline.

Ekumene Lysonge16:39
Well, I think I mentioned earlier, right? We're we're a very test heavy culture. We do a lot of A B testing. We're always trying to refine the the experience for the c customer and and leverage technology and other platforms we have, infrastructure we have, to present the best possible product to consumers. And so how that's shown up is, you know, in in in most cases, consumers today are are doing dis they're discovering financial products. you know, still through the traditional kind of SEO, Google kind of infrastructure or or experience, but more and more turning to these these foundation models and their platforms to do discovery, right? And so we have to meet the customer where they are, and that means that we have to reimagine in some ways nerdwallet in the way in which we we share and furnish information to the public. So we're we're figuring that out. but but what the claws and the the chat GPTs and the Geminis of the world perplexities and you can go on and on. What that's also done is allow development to dynamically or exponentially increase. And now you anyone in the company can vibe code an idea and within minutes and can be like an actual experience and a presentation that could be you know furnished to a customer. And so when you have the explosion of ideation, you have to also have a scaled governance model that supports the level of creativity that exists and and supports the possibilities of what could be you know released and shipped to customers and and I think that's what every person in my role is is is is managing I guess is the best word is how do I embrace innovation and creativity and ideation while also doing that in a in a way in which you make it durable and you protect the customer, you protect the company, you ultimately protect the board as well as their fiduciaries, right? To shareholders in terms of how they go about overseeing the governance of the company. And so that's not an easy job, right? Because you've got in in our case, 600 plus folks that are nerds, right? They're they're genius, they're they're you know very articulate and very accomplished. professionals in their various disciplines and they have all these ideas and now you have a tools and the ability to bring those ideas to fruition and and present them to the customers almost in real time. and you gotta have the governance in place to one enable that, allow for that, but also to do it in a way that's durable and doesn't wreck the company. And I will be the first one to to admit I don't have all the solutions because the the the the questions change every day, right? And so the answers change in in a very dynamic fashion as as the questions are presented. Well I think you have to be the the I think you you mentioned earlier, how do you get upstream? I think you get upstream by being but by showing a willingness to engage in the conversation around business principles and how we achieve the goal that we set out. and then you have to also have some flexibility to To fail. We're gonna make some mistakes along the way, but you just have to make sure that the failures and the mistakes aren't catastrophic to the business. And and the only way that you can ensure that is by being a part of the process. I think a lot of times legal wants to stand on the outside and they just wanna give the, well, here's the r here's the the risk matrix, right? These are the these these are the, you know, the parade of horribles that could happen. And you make the decision business. I've encouraged my team, both on the legal and on the the safety side and security, be a part of the discussion, be in the room, you know, say, hey, I see where we're trying to go. This is great. I'm all in. But have we considered this? But have we considered that? Can we get alignment w from this perspective or based upon this consideration? be a part of the solution and not a part of. the the problem identification, if you will. And I think that's how you get in the room more and you get brought in earlier versus finding out after the fact.

Cecilia Ziniti21:07
So on the the AI side, you s mentioned culture of experimentation. You mentioned there's some AI tools that AI is changing how customers experience the process of financial research. And obviously NerdWallet has both the human experts as well as all this content built up over, you know, more than a you know decade of of doing this. I I what it sounds to me like, okay, so a lot of your what your team may be doing is

Ekumene Lysonge21:21
Yes. Mm. Mm-hmm.

Cecilia Ziniti21:37
We've talked on the podcast before about product counseling. So getting in there in the room where the experience to be launched is being described and taking that yes and approach or the yes, but let's figure out these things approach. So I I wanna I wanna use a real example because I think it would be fun for people that that are in a similar position and and look up to you and your team for how you do this.

Ekumene Lysonge21:40
Yes. Mm-hmm. Yes. Mm-hmm.

Cecilia Ziniti22:03
But but I don't wanna I don't wanna have to use like a nerd wallet example. So let's just pick this one. So in Canada last year, there was an airline, and I think it was Air Canada, and their chat bot, it was one of these classic customer service chat bots, you know, powered by AI. But essentially the person said, Hey, you know, I want to get a bereavement fare. But what happened was they asked for the bereavement fare after having taken the flight. And that was not the policies that you had to ask in advance.

Ekumene Lysonge22:13
Mm-hmm.

Cecilia Ziniti22:32
And the chat bot said, Yeah, we offer a bereavement fare. Here it is, you can get the discount. So it just basically either hallucinated or, anyways, it got it wrong. And the court actually held, like, hey, in this case, you know what, even though it linked out to the proper policy, you're you're you the airline in that case was ordered to pay the discount. Now, who knows where the law has gone since, but something like that, walk us through you're being proposed you know, a chat bot or you're testing it or what have you.

Ekumene Lysonge22:41
Mm-hmm. Mm-hmm.

Cecilia Ziniti23:02
your team is advising on something like that. And I don't know how real of an example that is for you. But it's like all right. So yeah, how how how how how do you advise on that? And then is it really a legal issue? It's kind of to your point, maybe it's a business question. But anyways, how how would you advise on that? you're in that meeting and you're

Ekumene Lysonge23:07
No, I think that's a great example though. Right. What yeah, yeah, I don't think that either one of those are are necessarily business. That that to me is a a customer service and a policy driven kind of consideration, right? but now let's think about it. So, you know, in in our organization, I have lawyers that are embedded product counsel that they that truly serve as like quasi GCs of their re respective business unit or operating company. And so, you know, in you know, the deployment of this chat bot. What I would hope in in you know using the Nerdwallet wallet model is that product council is in the room, they understand the the the goal, right? The end result that we want to present to the customer, which is a a seamless kind of chat experience that would rival what they would have with a a normal customer service representative, right? Understanding that the customer service representative in many cases has some discretion to make some judgment calls as they're dynamically engaging with customers. So we have to, with that foresight, understand that probably the AI chatbot needs to have some degree of discretion as well. Right. Now, most people would say, whoa, whoa, whoa, we're giving computers discretion. But Remember, the goal is for the customer to feel like they're dealing with a human being. And the human being has some discretion, which means that you have to build into it the discretion for the AI to make some judgment calls as well. That's gonna be scary to people, but ultimately you have to be found foundationally aligned with the business outcome that you're trying to achieve, right? And so with that said, right, what I would hope if using a nerd wallet example is that the b My embedded lawyer would be in the room, say, let me see the the the the user interface and how how it's presented to the customer. And then ultimately, can we test it? Can can I be a part of the sandbox to test it against a number of conclusions? And what I would hope that my lawyers would do is to go to leverage AI to ask, give me all as many. Organic natural questions that a user might ask to this chat bot so that I can test. And then go in the sandbox and ideate with the team on, hey, have you thought about this? Have we thought about this? But if this happens, what happens? So that you can cover off 85% of the use cases that might be out there, understanding that we have to build in some flexibility for the 15% we didn't think about, but ultimately still steep everything in. presenting the best and the most realistic customer experience to the customer. So I don't think that that situation that we just went over with Air Canada should have ever gone to court because ultimately I would imagine that the goal was to deliver to a customer the most realistic experience. And what will we normally do for a customer? If they got access to a s a benefit that they otherwise had to Get pre-authorized, wouldn't we honor it? Because that's the right customer service thing to do. And so you kind of get to the right, the right conclusion by just use I hate to not saying that anyone in this situation didn't use common sense. I'm a big person on common sense, like common sense and being meeting the customer where they are. And ultimately, if it's going to build a customer that we have for life that will come back and return, because we made provision for him or her.

Cecilia Ziniti26:30
Yeah. Yeah, yeah, yeah.

Ekumene Lysonge26:54
during this very, very rough time in their life, we need to do right by the customer because we're in the business of having recurring customers. So I hope that I hope that was is responsive. Yeah.

Cecilia Ziniti27:02
Yeah, no, I mean that's great. And I think it makes sense. And then you're like, okay, you're litigating this, and like the press that you got over it is probably worse than the original thing. But yeah, I mean, then you get the obviously ridiculous examples of the people trying to, you buy a Honda for 10 bucks or whatever through the chat bot. And like that one's ridiculous, but this one was legit. I think it's interesting. Like, we had a similar conversation with the general counsel out of South Africa running a large healthcare system there, and he said,

Ekumene Lysonge27:10
Exactly. Yeah, yes. Yes.

Cecilia Ziniti27:30
In healthcare, the kind of like obviously people think patient first, but even on billing, that he essentially created something where there could be concessions on billing at the lower levels, even notwithstanding that, like, okay, is this an admission? Is this whatever? And it's like, okay, you know, no. And he and he felt that that was or he explained that that was relatively revolutionary, but that similar where in this case it's a private healthcare system and they they've done very well in part because of that customer centric.

Ekumene Lysonge27:39
Mm-hmm. Yeah.

Cecilia Ziniti27:57
approach. No, I I I love that. I love that. All right. So product console.

Ekumene Lysonge27:57
Yeah. Yeah, and putting the customer first, putting the customer first will lead you sometimes to conclusions or outcomes that make you cringe maybe a bit as a lawyer. Like Yeah, yeah, yeah. And you have to get comfortable with that.

Cecilia Ziniti28:10
Yes, yes, exactly. Exactly. Yeah. But like getting over getting over the cringe. Yeah. No, the kind of like the well, and it's even, I mean, the thing is is like this is business like these trade-offs of like hold your nose and sign the contract. Sometimes you gotta do that. And I did I ran a reference for somebody recently, and the reference was like the COO to whom this person reported.

Ekumene Lysonge28:24
Yes.

Cecilia Ziniti28:33
And the COO said, you know, I said, what areas can they work on? And they said, well, you when he first joined, he was kind of like, he really didn't like holding his nose ever. It was kind of like, well, we gotta do this deal because it's gonna be fundamental for this new business or whatever. And, you know, I think like this eyes wide open of like, okay, this deal is not great for XYZ reasons, but the realization that, like, okay, the alternatives like are not great either. And this is like, you know, you you get in and you believe, and ultimately, like, Working together with the business, that's what it is. And presumably, in like in the case of Nerdwallet, you've had this success over two decades building that trust. And it's like, okay, in that case, it was a construction company. But anyways, like the point of like it is something where in-house you have to sometimes realize that like the legal best answer is just not always gonna be the thing.

Ekumene Lysonge29:20
Yeah, and the and the legal best answer is just one of the many considerations. And you have to be aware and cognizant of this if it was only a legal consideration, this is what the likely outcome. But taking synthesizing all of the different outcomes together, what's the best business decision is kind of an amalgamation of all of those things, right? And it comes out to maybe seventy or sixty percent kind of z low risk and and maybe you know there's forty percent of it that's a little bit more risk than you might ordinarily be comfortable with. But no business has been successful and scaled without taking some amount of outside risk at some point in time. So

Cecilia Ziniti30:08
Yeah, no, exactly. I mean, it comes up a lot on the pod of like the most secure software is software that does nothing. Like, you know, so it's it's very secure, but it literally does nothing. So yeah, yeah, exactly. all right. So let's talk about AI a little bit. You said the future of in house legal is the legal team using AI to evolve from being a bottleneck into being an operating system for the company. that's an interesting take. Say more.

Ekumene Lysonge30:14
Yes. Exac exact exactly. Exactly. Mm-hmm. Yes. Yes. So, you know, from from my perspective, I I think you know early in the pod I mentioned I s I see kind of trust and safety as air traffic control function, right? And so similarly, you know, and I I can speak to Nerdwallet, but I've been at many companies over my history, you know, i in differing kind of capacities, legal has always been kind of a command control center at the companies that I've been in. And I think AI has kind of going back to an earlier comment, AI has enhanced the velocity in which people can ideate, create, and ship ideas. And so legal has to be able to evaluate, give give perspective, and ultimately advise at the same velocity. And there is absolutely no way to do that without leveraging some form of of AI AI tooling or in in in many senses a agentic solutions, right? I I think that that is a reality that in-house departments have got to fig figure out and solve for. We're doing that. We're we're figuring out how what what a agentic solutions we can bring to bear on things that are mundane and routine like contract review and product marketing review and things of that nature. And we've got some great partners out there that that we're leveraging to to help us solve that problem. But in order to for it legal to scale, or in my capac case, you know, for legal and security to scale, we we have to bring to bear some some truly advanced gentic and and AI solutions to to help us just keep pace with the work, right? And and so from a command center perspective, we have to also then orient the business into understanding that in some cases, for heightened, more judgment-based considerations, you're going to engage with a human, a lawyer. For things that are routine, kind of call it run-of-the-mill assembly line work, you're going to be dealing with an agent or a non-human-based solution. And that has a Different set of SLAs that has a different degree of error correction and and and imperfection to it. And we all have to be willing to move at this new pace and accept kind of a new experience in-house. And and I think a lot of our work at Nerdwile has been around: hey, explain it to the business, we're going to evolve dynamically. with the pace of technological kind of advancement, that means that the experience for the end user changes in-house, in from a cust constituent perspective, experience perspective. And we need everyone to give us grace as we figure that all out together, right? and we've done a pretty good job of that. I think you know the speed that we've been able to deploy certain solutions has helped people move things through. through the w the the warehouse, if you will, to to ultimately ship it. but also that that's come with quite a few pivots as we learn more. We we understand how we need to change certain things to to be more corrective in terms of what it ultimately delivers to a requester, whether it be you know a a request for a contract review that has an agentic s an agentic agent associated with the the initial pass of that review or in some cases the routing of approvals and and signatures and that don't have a human being involved with it. it's it's all for me a very exciting time. Like to to figure this out in real time while the ground is shifting beneath me. To me that's exciting, right? It it's for someone that's been practicing, you know, 25 or whatever year, 25, almost 26 years. These little moments in history to live I'm living history, right? We're all living history. is exciting. So

Cecilia Ziniti34:51
Yeah. Yeah. Yeah. I mean, I remember on the Living History Point, the first time I used Google back in two thousand and I was like, wow. And same thing, first thing you time using AI and now first time having an agent end to end review a contract.

Ekumene Lysonge35:03
Yeah.

Cecilia Ziniti35:08
Yeah, so so tactically you it sounds like you've asked for grace from the business as you implement the agentic things in legal. so how are you picking the use cases? And then if I'm a G C diving into AI, what are kind of your top three tips on implementation?

Ekumene Lysonge35:13
Yeah. Yes. Yep. So first, how am I picking the use cases? Obviously, you know, I I think I mentioned earlier you you try to do things that aren't gonna be catastrophic if they go horribly wrong, right? And so you you pick kind of what I'll call low risk kind kind of items and you you you test around those things. So, you know, vendor agreements that are below a certain threshold. Do I care if the widget contract for a hundred K is wrong or something like no, I don't nobody cares. you know, or an NDA, obviously you have guardrails. We've been practicing with these guardrails and who knows how many and countless NDAs we've signed over the course of the last you know four or five years. you start with the things that are routine and and mundane and low and minimal risk, and you test, you use the I don't think lawyers believe in testing a lot of things. We want everything to be perfect, that's part of our problems, right? Is that we are we are the the s kind of slaves to perfection, if you will. And I think for somebody like you and I who have been doing this a long time, I'm very comfortable with something being about 70%, 70 to 75%, because most of the time I only get a chance to come up with advice that's about 70 to 75% accurate, right? and so you know, I'm comfortable living in that space, but again, a as practitioners, I think we have to be generally comfortable with saying that what I'm going to roll out here in this kind of AI powered world may only be about 75% or 70 75% accurate. And I've got to give myself grace and I've got to be willing to to operate in the the the ambiguous or nebulous world of what could what could potentially happen and be agile enough to respond accordingly. So that that's kind of the answer to the first part of the question. I think the second part is how would I what would I advise? I would advise, you know, if you don't have a legal operations function, think about it. I think, you know, legal operations and the legal prof operations professionals are becoming a a a an absolute requirement for anyone that has wants to have a scaled and and resilient kind of legal in-house legal function, period, full stop. And so it having a a very seasoned legal operations team or a seasoned legal operations professional helps you sit down, go through all of the different work streams that come through the department, evaluate what are the the routine, mundane, repeat repetitive, kind of monotonous things that we do and synthesize that into a list and then take that list and say, these are the things that are. give give way to kind of some form of AI powered agentic solution, right? And that's what we did. We went through, we evaluated all of our different work streams, figured out what do we do on a routine basis that, you know, if if we had the s tech tech technoc technological solution to to solve for, we would And then we took that list and we said, okay, this is where we're going to start. Let's knock out how we can bring a s an AI-powered solution to dealing with these tasks. And I'm I'm proud to say we're probably three quarters of the way through that list. and and that has freed up my legal professionals to then focus on the things that matter most, which is bringing their judgment and expertise to bear on the most essential problems that are being confr that the business is being confronted with for which they need you know, legal advice to to s solve for and answer. So that that's how we've gone.

Cecilia Ziniti39:29
Awesome. What's been the most successful part of the rollout?

Ekumene Lysonge39:33
successful part of the rollout. Well, I I mean, I don't know for you. I've s probably many of your listeners. There's certain the older I get, the the more small things seem tend to annoy me. And so so so so the so like the the this for forever, you know, we have to triage so much. And and even myself

Cecilia Ziniti39:50
Ha ha

Ekumene Lysonge39:59
You'd be surprised by the number of slack messages I get from nerds about how do I, who do I go to for this thing? Or it doesn't matter how much publicly available information there is about who to go to for a thing, or what do I is this approved, or can you a lot of it is forum shopping and things of that nature, which we won't get into here, but but but that kind of over time on a daily basis annoys me. And so forever I've been trying to figure out. How can we bring an ask me anything experience to a channel Slack, right? That allows for all that triage and all those frequently asked questions to be routed to an agent. Have somebody else give you that answer, right? small for some, huge for E. But at any rate, I I'm proud to say that we are now in the final stages of being able to release that, which again will free up if you. Multiply if if say for example E is getting and this is not an overstatement I might get 30 of those questions a day via Slack, right? Multiply that by I've got 17 people in legal and another 17 people in security, and they're probably getting 20 or 30 of their own questions, right? That by solving for that and having an agentic solution for that.

Cecilia Ziniti41:05
Mm.

Ekumene Lysonge41:23
that frees up capacity for the entire team that then gives the team the ability to really focus on much more substantive kind of topics and and questions and render, you know, even more substantive advice that it that moves the business forward. And so I'm very happy about finding the right partner to to help us with the s solve for that thing and being in the final stages of deploying a solution there. So

Cecilia Ziniti41:51
I love that. All right. Let's go to the lightning round. What is a book, podcast, or mentor that shaped how you think?

Ekumene Lysonge41:57
Lightning round, okay. well I have a min several mentors. kind of the the one that that's that stands out is I had a a mentor, his name his name is Ron Brown. He's he's the general counsel of a real estate holding company, a conglomerate of different companies in Indiana. and called Pedcor. And it when I worked for him We would, you know, we would go on these long walks around. There was a reflecting pond. The story is gonna it's gonna meander a little bit, but he he smoked a lot of cigarettes at the time. He doesn't smoke anymore. But so he would always want to smoke, buddy. So we go outside and we walk around this reflecting pool while he smoked a cigarette and we would really just talk about legal problems and things we needed to solve, right? And I in at that time in my career, you know, I I wasn't

Cecilia Ziniti42:39
It it goes, go, it's good.

Ekumene Lysonge43:01
I was transitioning into being more of a high level executive kind of thinker, right? from a pr legal perspective. And we would talk and and talk and I would give him all these super technical things. And he was like, E, let's let no, let's let's just break it down to basics. Like, let's just talk about this like a normal human problem and figure out how to like like what's the common sense solution? Like, okay, put the legal stuff, put the as it like he was put the legal shit aside and like let's

Cecilia Ziniti43:26
Mm-hmm.

Ekumene Lysonge43:30
Talk about it from like a common sense perspective. And then we would just have a a more organic conversation about what's the most sensible, pragmatic, kind of common sense way to solve a particular problem. And that resonated for me. Why at that time, I don't know, but it what it helped me reorient myself toward is not thinking I have to over-engineer solutions or over-engineer advice. Sometimes it's as simple as just being a voice of reason and pragmatism and common sense in a room. It's amazing to me the higher I go in in corporate America, the the the less common sense is is brought to bear on things. I think because people feel like because they've got so much experience and because they've seen so many things and because you're working in a corporate environment that the answer has to be complex. A lot of times it just has to be kind of a common sense thing that's steeped in a first principle, right? And so that that's the the I think about I I kind of hearken back to those times of walking around the reflecting pool while he smoked his cigarette and the kind of just the general things that I learned from those those talks. And I pull more from those conversations and the the problem kind of the the solution kind of process. that we went through there than probably other times in my career which where the stakes were higher and it probably more sophisticated or technical kind of application of principles were were what were at stake or were n necessary to to kind of talk about the solution. But yeah. So

Cecilia Ziniti45:15
love that. I love that. All right. See. someone pick what would you like for folks to take away from our conversation?

Ekumene Lysonge45:26
I I think, you know, for me, I think you you you spoke to some of the things I've written about and this whole notion of of decision discipline, right? And it and it's not necessarily I think more I would encourage more of my peers and more of us in in the pr the in-house practice of law to to not be afraid of change and the dynamic shift that is occurring in the practice of law, right? Because of AI, right? I think that if anything, for those that have been practitioners, this is a doorway for what it could be infinite job security. From from my perspective. And what I mean by that is by being the person in the room that brings rigor to the discussion and Is there to evaluate the decision, help evaluate the gaps, have the institutional knowledge of the failures and where we've missed before, and bring that to bear in the conversation and to be air traffic controlled, to make sure things are always aligned around a set of common perspectives is an opportunity for us because for so long we've wanted to be at the table. And I think because of the nature of AI, it gives us. in automatic seat at the table, really in perpetuity, if we don't mess it up. Right. And so I I think I think that bringing the kind of the decision discipline to bear on conversations, whether it be at the board level, helping to to to facilitate postmortems on decisions that are made, to help be a voice of reason to say, we did this before, here were the gaps that we had. This is how it be came to bear on the ultimate performance of the business against whatever the initiative initiative was is a a value add that I think may be lost if we don't if we don't become the champions of of of bringing that form of of conversation or rigor to these strategic discussions. So I'm really excited about where we are as a profession in the future and the opportunity for us to to be vo the voice of reason and and discipline in the room and I hope other people embrace this kind of dynamic or fundamental shift as well and think about it from that perspective. I'm a glass half fool kind of person. So that's the way I that's the way I look at life.

Cecilia Ziniti48:06
Let's go. Thank you. I love that. Ekumene, that was this was a really important conversation. Thank you so much for being here.

Ekumene Lysonge48:14
Yes, thank you. I really appreciate it. Take care.

Cecilia Ziniti48:16
Awesome. That was Ekumene Lysonge, Chief Legal Officer and Corporate Secretary at NerdWallet, and one of the sharpest thinkers we've had on the show on legal leadership and where we're heading. Whether you're leading a legal team, looking for infinite AI job security, building legal tools or working with clients in-house, head over to gc.ai to see how the most forward-thinking folks like E are doing their best work. Follow CZ and Friends wherever you get your podcast. Thanks and we'll see you next time.