Legalweek 2024: Generative AI, from Innovation to Impact

Episode Notes

Transcript

Ed Watts :

Hello and welcome to another edition of On the Road with Legal Talk Network. I’m Ed Watts and I’m the host for this episode, which is being recorded on location at Legal Week 2024 in New York City. With me today, I have Jamie Brown from Lighthouse and Isha Marathe from ALM. Welcome.

Isha Marathe:

Thank you so much. Thank you for having us.

Ed Watts :

Fantastic. So looking forward to getting stuck into this conversation. First of all, Jamie, I’d love for you to explain what is Gen ai?

Jamie Brown:

Well, generative AI is a type of artificial intelligence that can create new content using data and algorithm. So at its simplest, that’s the simplest, easiest explanation. It learns from existing examples of other human made content that content is content that is accessible and it can be used for lots of different purposes. Common examples are to enhance creativity, discovering new insights, or something as simple as automating tasks. The best example of generative AI is chat GBT, which was released by OpenAI November of 2022.

Ed Watts :

Yeah, and how does it differ from AI of SDA AI has been in round for five plus years? What’s prompted this new phase, this new buzz?

Jamie Brown:

I think it’s helpful to think of it in the context of how it fits in the broader AI universe, if you will. Gen AI is really just a subset of ai. It leverages machine learning and also deep learning algorithms ultimately forms large language models that train the technology to generate output with increased accuracy over time.

Ed Watts :

Fantastic. And Isha, how have you seen AI in the kind of publication space in the industry, the evolution? How are people viewing this change in access to technology?

Isha Marathe:

So I’m super new at Legal Tech reporting. I really started two years ago and I knew very little about it. So when I started it was all CLM and E-discovery and just learning what that meant and things weren’t changing that my entire first year, halfway into my second year, things weren’t changing that fast. We heard a little bit about the Metaverse, we heard a little bit about CLM was really hot last year or the year before. It was super hot, it was all anybody was talking about. And then gen AI happened at the end of 2022. And in my really short time reporting on it, there’s really nothing I’ve seen similar to it in the legal or I mean as a journalist in the media industry. And it’s really caused

Incredible innovation and incredible panic and I’ve not seen any tech in my lifetime really do that. And that’s my sort of general answer to that question. There’s a lot that’s changed in terms of the tech. There’s a lot that’s changed in terms of fears of people losing their jobs because the tech can really do so much. And I’ve demoed so many tools, eDiscovery tools, but also CLM tools and matter management tools and whatever jargon you want to use to describe them. And it really is super cool technology that can do things in a super user-friendly way that seemed really difficult and tedious and boring. So that’s what my answer would be, which is, I dunno, I hope I answered that.

Ed Watts :

So rather than being fearful or maybe resistant to this technology, how do you think people should be seeing this and perhaps embracing it?

Jamie Brown:

I think it’s really an opportunity to harness technology to improve day-to-Day business process. That’s really what is getting most businesses excited about. I think for the type of clients that we have in law and technology, it’s also an opportunity for those folks to partner with their peers and their organizations to help them enable it in a way that drives the business and the innovation and the things that we were talking about. But also does that in a responsible and defensible manner. There are risks that do require mitigation, and in order to be in a position to mitigate those risks, you have to understand a little bit about how the technology works and you have to understand a little bit about how the business contemplates using that technology and really be in a position to advise the business or your corporation on the best way to proceed. Because there’s a lot of nuance depending on the use cases and again, how the business really wants to deploy it.

Ed Watts :

Yeah, and really interesting. Do you think there’s a formula that people should be applying for adopting AI in their organization? How should leaders, managers, owners, CEOs be thinking about implementing AI into their organization?

Jamie Brown:

It’s a great question. I guess the short answer would be that it really comes down to gen AI governance. So like the deployment of really any other transformative technology, you really need to have some kind of governance committee have a good understanding of your mission, how are you going to deploy it and really put the time and effort into planning. That doesn’t have to be exhaustive, it doesn’t have to block deployment fully, but it really does require the right stakeholders that do include legal and security and and privacy and other business stakeholders to really again, define those use cases, have some preliminary understanding about how the technology works, evaluate really case by case. I mean use case by use case, evaluate the risks that are associated with a deployment because it is going to depend and try to mitigate those risks. And then once one is ready to deploy the technology, really doing it with some guardrails in place, making sure there’s documentation, making sure there’s training. And the biggest piece is that this really never ends. So you have to have some kind of evergreen maintenance in place because the technology is only going to continue to improve and evolve, but the adoption is also going to continue to evolve and expand. And so that governance really never goes away.

Ed Watts :

Yeah. Okay. So would you recommend that people realize that this is kind of Pandora’s box once you open it in your organization? There’s no turning back.

Jamie Brown:

Yeah, a hundred percent, but I would say I hear the term Pandora’s box and it sort of suggests a little bit more of the fearmongering that I think is really happening. It it’s powerful technology that presents massive opportunity. That’s why it’s innovative, that’s why it’s transformative. And what you were saying before is really accurate, which is we don’t have this come by all that often. It’s not every year that you have really innovative transformative technology that we’re talking about sometimes. I’ve been in this space for a really long time going over I think 23 years now. I’d say every five to seven years there’s something really exciting. So five to seven years ago, we were super excited about cloud deployment and what cloud transformation means, and we’re still excited about those things and they’re still quite relevant, but this is really that next wave of innovation and excitement. But along with that, sure, it comes some risk and that gives lawyers and legal practitioners a huge opportunity to really take the skills they already have and port them into a new area that really is just so ripe for education and advice and real partnership to help the business drive this technology forward in a way really that’s contemplated and that really shouldn’t be scary. It should be exciting.

Ed Watts :

Fantastic. I think we’ve seen examples of innovative and forward thinking companies encouraging the users of Gen AI and the organization in a way that’s not prescribed to think of new use cases to push the boundaries. And we talked about AI being evergreen and needing constant compliance and checks and balances. How would you recommend that people keep on top of it, they’re able to understand how people are using AI that’s already been deployed or maybe not sanctioned in a compliant manner?

Jamie Brown:

Yeah, so I would say for, if we’re talking really about a legal stakeholder, it’s really letting those other stakeholders know that they deserve or need a seat at that table so that they have early insight and perhaps can contribute to some of those use cases to make sure, again, that they’re appropriate, that they’re being ruled out in a way that’s thoughtful and really responsible and defensible. The second is that one of the pieces of advice that we give customers is to make sure that they have an ongoing inventory of AI use cases within their organization and that those are continuing to be evaluated. And what’s interesting, I think about the technology is just sometimes even reading about those use cases really opens one’s eyes around just the power. And I would say the third would be for individuals regardless of really where they sit in an organization to take it upon themselves, to educate themselves about how the tech works. There’s a lot out there, even right now, people are getting educated in a bit of an organic way because there’s just such a demand and a hunger for knowledge, but getting educated through their peers sometimes, or they’re partners, it could be a vendor, it could be a law firm, and really kind of talking through that. I think they can bring that knowledge back to their organizations to again partner with them on successful defensible responsible deployment. Yeah,

Ed Watts :

Awesome. So Gen ai, prompt prompts are a big topic at the moment, how to ask the right question to get the right answer. Is there a chance that two people wanting the same thing can ask their question in two different ways and get two different responses?

Jamie Brown:

Sure. There’s a whole, almost like a cottage need, if you will, around training on prompt creation. It’s almost similar to how one would train perhaps an associate on how to take a deposition. What is the difference between perhaps an open-ended question or a leading question? How do you remove bias perhaps from certain questions? So prompt creation is really important and understanding, I think, how the technology works can help inform that a bit. The other piece that’s really important is really validating the results. And once you go through enough exercises where you are validating, even that process will over time improve how you answer questions. Yeah.

Ed Watts :

Awesome. Fast forward five years, where do you see Gen ai, AI in the workplace, AI in your personal lives? What does the future look like?

Jamie Brown:

The power that this has is really to transform business process really across almost every single business unit. So from a business standpoint, I really see being commonly used, widely used to support the way that people work. And if you think about what that means for e-discovery professionals, it really means that they will have the ability to readily preserve and collect information that is generated by this technology, which includes prompts and references to documents and new content and creation. So that becomes really an integral part of their process to the extent that information, of course, is relevant to that particular matter. I also see other use cases within e-discovery where we are actively using prompts perhaps to support the discovery, the creation of discovery requests or responses, or create documents that perhaps summarize events or timelines or help us understand perhaps who key actors are or their roles in a particular investigation.

So I continue to see an evolved use even by eDiscovery. If you think of legal or eDiscovery as its own little mini business unit, from a legal and regulatory standpoint, the law is always a little slower than the technology. So I do think we will continue to see a really mixed and evolving landscape just like we are now. It causes a lot of confusion. It makes it a little difficult to feel as though you’re operating in a fully compliant manner, but I think many of us will look to some of the guidelines that have already been promulgated, particularly those in Europe, like the AI Act as really setting a standard for us to follow and adapt, where we’re just continuing to implement and evolve the use of this in a way that meets guidelines, that is responsible, that is aware. And then lastly, I think that we will see companies in a position where they will need to perhaps defend their use of ai, and the only way that one can defend it is really if you have that documentation in place, if you had that AI governance in place and really are able to substantiate not only a defensible deployment, but really maintenance and evergreen maintenance of that program over time and an awareness, and much like companies have to do with really any other compliance response that they may undertake.

Ed Watts :

Yeah. Fantastic. Super interesting. I’ll ask both of you this question, but first, Isha AI overhyped under hyped or just right?

Isha Marathe:

Well, I guess when we talk about ai, we have to specify what type of AI we’re talking about, because AI is not new necessarily, but if we’re talking about generative ai, I don’t think it’s overhyped. I think that some people who created understand it better than other people who create it, and I think sometimes there’s a lot that gets lost in marketing hype. Maybe the hype is in the marketing, but not in the technology in this case, I don’t think so.

Jamie Brown:

Think if we’re talking about gen ai, I would agree that this is about just, right, certainly as it relates to the potential for disruption and innovation and productivity improvement, I think as it relates to legal and the risk associated with the use of it, I also would say it’s just right with the caveat that you do need to have that governance in place. If you have the governance in place, do have the partnership, you are planning, you’re doing this in a responsible way. There are ways to mitigate that risk, and sometimes that part of the message gets left out

Ed Watts :

Well. It’s like we’ve reached the end of the road for this episode. I want to thank our guests, Isha and Jamie, thank you so much for joining us today. Thank you.

Jamie Brown:

Appreciate it. Thank you so much for having us. This was great.

Ed Watts :

Also, thank you to our listeners for tuning in. If you like what you heard, please wait and review us in the Apple Podcast, Spotify, Amazon music, or your favorite podcasting app. Until next time, I’m Ed Watts and you’ve been listening to On the Road with Legal Talk Network.

Speaker 4:

If you’d like more information about what you’ve heard today, please visit legal talk network.com. Subscribe Via, iTunes and RSS. Find us on Twitter and Facebook, or download our free Legal Talk Network app in Google Play and iTunes. The views expressed by the participants of this program are their own and do not represent the views of, nor are they endorsed by Legal Talk Network, its officers, directors, employees, agents, representatives, shareholders, and subsidiaries. None of the content should be considered legal advice. As always, consult a lawyer.