What are legal professionals actually using AI for right now? Listen in as one of the leading and well-known experts in the field talks about what is out there, how people are using it, and what you should be paying attention to.
We also discuss tips for success on LinkedIn, and describe many of the legal technology conferences Cat attends, specifically highlighting those that offer the most value for solo and small firm practitioners.
#LegalWeek2025 #Podcast #TechTalk #AIinLaw #LinkedInMarketing
Ryan Bankston:
You may have seen her on a webinar or listened to her on another podcast. Catherine Casey has a distinguished ediscovery career. She was once a director at KPMG and PwC and at Gibson Dunn before landing in the C suite at disco and now, several years on, at reveal brain space, which operates several well known tools, such as logical trial director cloths described as an AI baddie, and she's on the New York State bars AI task force that published recommendations on the ethical use of AI back in April. She does all the conferences. She knows all the people, and they know her Cat Casey.
Cat Casey:
Oh, thank you for having me. I'm going to blush now. Goodness, what an intro.
Ryan Bankston:
Your reputation certainly precedes you, and it's hard to miss you. If you're active on social media, if you're in ILTA, if you're a member of the LA legal tech in general, you're probably going to come across a cat Casey post or article or something.
Cat Casey:
Potentially my alter ego, techno cat. It's a lot about AI lawyers embracing it and a an egregious amount of sparkles.
Ryan Bankston:
We could all use some more of your published e discovery content so helpful. It's got such a long career behind you. There on E-discovery some distinguished positions. Could you offer this audience some of your practical tips for preservation and collection?
Cat Casey:
Yeah. I mean, I think one of the things that's changed the most in the almost 20 years I've been in ediscovery is kind of the volume, variety, velocity of data in that there are new platforms and tools emerging and dominating the market and potentially able to be completely banned, as in the case of Tiktok, and so it's more important than ever to understand all the applications that your organization is using. Right now, there's 80 to 90% of people using shadow IT, or my new favorite term, AI smug. So you need to understand what tools, applications and platforms are hosting data that's potentially, you know, subject to a preservation obligation. So map it, understand it, reevaluate it, have policies that, more than anything, is the number one thing, and have contingencies in place. I know a lot of us are moving to preserving in place, but as was the case with like tick tock, some of these smaller platforms get very big, dominant positioning can have a whole host of potentially data, and then can be pulled offline. So you need to have a way to quickly grab that data so that you don't have an exfoliation issue. So plan broadly, understand where your data lives, and don't just rely on preserving it in place, because you need to be able to kind of tactically react and grab data as you need it, and that's a little bit different than what I would have said a couple years ago. But I think with the consumerification of AI and other applications, there's an entire generation of us that grew up with this cell phone that has all these great, sexy tools, and we expect our business tools to be great and sexy too, which means, I think more people are embracing shadow IT alternative applications, which means the universe of data that's discoverable is much more widely spread and disparate, and we as practitioners need to know where those risks lie and know where the landmines are and help navigate it. I don't know if that answers the question, but that's kind of what's been top of mind for me.
Ryan Bankston:
Yeah, no, it's very helpful, especially some of the things you brought up. Like, what was that? First thing you said, AI smuggling.
Cat Casey:
No smuggling. Like, smuggling like a smuggler. Because what it is is people are trying just so they may have a policy about not using certain AI applications. But I mean, 25 years of being Pavlov to Google for me, was undone in about 25 days of using generative AI. I think a lot of people are creating exposure by using AI, which we should be, but not under the purview of their organizations. And so it's kind of beholden on organizations to not just say, Oh, we banned X, Y or Z application. You really have to look further, because I didn't know they were using it as never defensible in front of the judge.
Ryan Bankston:
Yeah, that makes sense. That's something I hear talked about a lot, especially when it comes to things like, which tools are actually safe, or how you should structure a query with a tool or a prompt, and you know, what kind of information should you include, or could you include? Like, what risk are you taking when you put something on a prompt?
Cat Casey:
My number one advice to anyone, and a lot of what I've been talking about is prompt engineering and embracing generative AI, I literally have been screaming about AI like. Pushing a boulder up a hill that's on, you know, boulders on fire falling back on me for 20 years. But the number one thing I tell people is, if you haven't read the terms and conditions, actually read them, because every tool out there is clearly telling you how and when they're capturing your data and how they're going to use it. And if the tool is free, you are the product. And that isn't to say don't use the tools, but you know, there have been cases around confidential information getting out from people putting proprietary code in, like Samsung. There have been cases of confidential information kind of being triangulated or being, you know, used to train the models, and so you've got to be really careful. There's even a concern about potentially vitiating privilege by putting information in Oh, and by the way, much like a Google search, it's not discoverable in every case, but it's still a discoverable thing that could actually be relevant in a case. So there's just, there's a lot of things to think about. You know, as you're navigating this really, truly exciting time, this AI Renaissance, I am here for it. I am a fan girl. But there's a lot of especially in legal things, to kind of bring our legal mindset to.
Ryan Bankston:
Yeah, that makes a lot of sense. Yeah, it's a crazy world out there, especially thinking about how there's a lot of opportunity to enhance our work when it comes to productivity and efficiency. But then in the legal space, there's also the opportunity created by people using tools badly and creating opportunities for plaintiffs attorneys. So like, I'm thinking about the McDonald's case that settled several months back because of BIPA in Illinois, where they had their AI analysis on drive throughs. So yeah, there's a lot of opportunities out there in this, this AI landscape.
Cat Casey:
It's all about balancing, like Fear and Desire, but also, much like a tsunami, we aren't going to stop it one molecule of water at a time, like the transformation is here and happening. Like, one thing I invite people to is we try to compare this AI moment to other things, like maybe to the internet, or to the printing press or the written word, and it's not really like that. It took 20 years to get 100 million users of the internet, and they were still publishing that the internet is dead 19 years in, it took 10 years to get 100 million Netflix users. It took 65 days for one application of Gen AI to get 100 million users. It's much more ubiquitous and moving at a much faster speed, both in terms of adoption of consumer grade, which has kind of been the barrier of lawyers embracing it. They're like, Oh, I don't want to explain this new foreign thing. Well, if your nephew, your son, your client, your brother, your neighbor, all using AI, those barriers are gone. So it's much more ubiquitous, but also just the speed of the models changing. We've had 678, open AI models drop in the last 18 months. We've got deep seek. We've got all these other models that are dropping. Oh, and there's an entire revolution around AI agents who can act on your behalf. So they just things are moving very quickly and trying to not embrace AI is backwards looking and will hurt you in the long run. You kind of have to embrace it. It's just about finding that safe, defensible, ethical way of doing it.
Ryan Bankston:
I've seen a lot of attorneys that are, you know, they're pretty eager to use it in areas like market, you know, or on their website, for articles, social media and things like that. But when it comes to using it in their word product or even in eDiscovery, be a little bit more timid. Understandably. What could you speak to as far as what generative AI tools are available in eDiscovery, and how are folks actually using them?
Cat Casey:
Yeah, so the interesting thing is, because the plane isn't already built. People are building it a little bit differently for each major technology company. So one of the first kinds of shots across the bow was a kind of AI chat bots. You can ask questions in free form, like you would to Google, but conversationally and interrogate your data source using AI. You know, who's talking about fraud in the data set? You know, what is the name of the person with this title? Is there any discussion of XYZ? So that's kind of the generative AI powered chatbots and or Gen AI powered search. So the chatbot is more of a helper. The Gen I powered search sort of thinks of it like, Well, no, don't think of it like this. It's not like search terms on steroids, because it's not relying on what you already know about the data set. It's letting you ask questions to find out what you don't know. And then the next piece is Gen AI powered review. One flavor is basically using Gen AI to kind of do the seed set like you would with predictive code, not predictive coding. Scratch that tar 1.0 and 2.0 there's also, which was what we're developing to reveal Gen AI powered search that can almost function like another, you know, review, or another person in the universe that's operating in parallel. To you, where the magic happens with all of these, whether you're doing the chat box. That the search optimization or the automated review is that ideally, it shouldn't be operating in a vacuum. It shouldn't be that I have my workflow, my AI workflow, my gen AI workflow. That's stupid. What it should be is I have my workflow. Here's where machine learning plays. Here's where data visualization plays in you know, I can take the questions from like our tools called Ask, from ask, and then I with the answers. I can look at the actual data at the documents itself. I can push that into, say, the cluster wheel, which shows me the concepts that are part of the answer to my question. So being able to have Gen AI play with this battle tested, already court approved, machine learning and more legacy AI is where the magic happens in my mind, because you're not starting at zero, you're building on the mountain we've already been building for 20 years. So that's kind of where I see it right now. And I think ideally, what we're going to keep seeing is not, people aren't going to stop at say, AI review, because I think adoption on that will be slower, because it feels a little bit more foreign, but that we're going to keep finding other little ways to either identify things Gen AI is better at than our existing tools, or find completely new things we hadn't thought of using AI to uncover, you know, summarization, connecting the dots and the data a little bit differently, making suggestions for how you even create the prompt. There's a lot of different ways that, you know, AI can work with Gen AI to make the end result of I have to look at less documents to get my answers happen. And I think that's what we're working towards. Not so much, can I look at every document more quickly, but can I get to the answers in a way that I can articulate to the judge that is defensible and repeatable and statistically valid without it costing as much time and money. So that's kind of the universe as I see it now.
Ryan Bankston:
Yeah, that's awesome. You mentioned, like existing AI tools. Are there any non-generative AI tools that are available that you think are very much under utilized?
Cat Casey:
All of them. And so you mentioned, you mentioned, back in the day that I was the global director of technology, I think it was for Gibson Dunn for several years. And one of my crowning achievements is I brought AI and I brought brain space in. But my adoption rate was still very, very low for even things like email threading, let alone just basic visualization. So my hope and I sometimes get eyebrows raised, sometimes my own eyebrow raises over it, my hope is that Gen AI is the gateway drug to use these underutilized tools, which are everything from organizational things like, I mean, email threading uses AI. People don't think that, but basic things like that, to anomaly detection. So using AI to see data points that stand out and shouldn't be there. To sentiment analysis, you know, identifying aggression or moods that maybe are aligned with bad behavior or high risk. To visualizations like concept clustering, which is identifying key concepts and bringing them together, or social network analysis, which is communication maps, who's speaking with who with what frequency, or just basic dashboards that kind of present all that together. And then even further along, we've got things like reveal has an AI model library. I'm sure relativity has something similar, where there's off the shelf models you can plug in that are trained on a big universe of data to do certain types of cases, you know, find privilege or find potential fraud or harassing behavior to even the next step is having AI models that are powered by non generative AI that are bespoke to your data set that you can build with our data science team, with whoever, and then carry it on to other cases, and then you, as the client, own the IP. So across this is a pretty wide spectrum, and then where the magic happens to my mind is okay. I can use Gen AI search to ask questions, reduce the data line, then push it into all of these universe things and same thing, even with automated review, like if you're using Gen AI to review and parallel a more effective way, you can use all those other visualizations and AI to reduce the universe of data that requires review to begin with. I mean, it's exciting, like, I think it's we're finally at the point where, here's how I like describe it, Gen AI is AI so simple, so intuitive, a lawyer will actually use it as historically well, like historically, we had all this great stuff, but a lot of times it would be a legal technologist, which is awesome, but I think where the adoption kind of falters is a lot of times the the key members of the case team with the case knowledge are hesitant to deal with stuff that feels really technical or like a relational database. And so having this kind of way to just write sentence typos and all free form and ask questions about the data, like a lot of the Gen AI search or the chat bots do, that's something even a partner would be comfortable doing. It doesn't feel like the barrier of entry to gaining all of these great insights is being lowered by having consumers. More human centric approaches to building AI, building AI, that is, it doesn't feel like AI, it just feels like I'm getting better answers faster.
Ryan Bankston:
I'm interested in where it really empowers the smaller teams to do more work. You know, like some of the things you posted about, reviews that were, you know, 9 million documents and you like, got them down to 50,000 or something. Like, it's like, huge, you know.
Cat Casey:
I think they're the ones who can benefit the absolute most, because, not just because of the David and Goliath, yes, that's part of it, but because there's not necessarily bureaucracy or countervailing pressures that don't benefit from being more efficient if you are small and lean and mean, you know, litigation boutique, a small plaintiff firm, you know, being able to get to those answers more quickly. You're not thinking, Oh, I don't get to push through this cost for 100 reviewers. You're thinking, I didn't have money for 100 reviewers, I wouldn't have found this information any other way. This is awesome. So I think we'll probably see more adoption with small and mid sized firms that are using it as a way to level up, but also they don't have as much to push back against. They don't, they can, they can get adoption more quickly, because their option is to not find the information otherwise.
Ryan Bankston:
I wanted to circle back, we were talking about preservation earlier, and I wanted to ask you, what is forensically sound data you were talking about this recently in some of the articles that came out with Delta.
Cat Casey:
You know, it really varies. I mean, I think that having someone that can attest to using the right tools and technology with the appropriate, you know, and see, type certification. That's part of it, being able to capture data at the moment in time, as opposed to something that may have had the metadata modified over time, especially when we're dealing with a lot of ephemeral messaging or continuous communication, becomes much more challenging. And then I think it's, frankly, it's not my super area of expertise, so I may push back a little on going too in the weeds of it, but it's something that's much more complicated than we have established it over the last 20 years, because the nature of data itself is much more mobile, adaptable and ephemeral in nature, and so forensically sound requires more moment in time capture and additional delving into, you know, metadata that we may not have done for more traditional, linear four pages of the four corners of the document, sort of information.
Ryan Bankston:
I was recently talking with someone that was commenting on how it's really hard to have, like, ethical walls in Microsoft office 365 and they were like, yeah, it's because they don't, like, it doesn't have an audit trail, like, you know, like, I manage your net documents. And I was thinking, well, actually, it does not everybody really has access to, like, you have to go into purview, or you have to have, like, a certain, you know, role that has to be assigned to to be able to, like, use the eDiscovery and purview? Well, you have a license for it. You have to have paid, yeah, right, right. But it does like, I mean, everything's not everything, but almost everything is tracked in some way, like that, metadata, you know, or in a certain way in Office 365.
Cat Casey:
Or it's tracked than people realize, even like purview in particular, if you're using some of their Gen AI agents to kind of collaborate on material, if you interact with other data that's in the universe of material that purview has access to, you may both be modifying and creating an additional record that is then stored in a totally different place. Because it's the generative AI functionality rather than traditional information. There's almost like there's parallel places information is being stored, and custodians, who may not have been given user access rights to certain pieces of material, if it's accessed by purview and they build upon it, may suddenly, then have an additional copy of the information, so it's trackable, it's audible. It's also more complex than people realize, as they start leveraging tools they maybe don't have as clear an understanding of where the data is living and what is being captured.
Ryan Bankston:
Right. Like how the co pilot follows teenage retention policies? Yeah, that makes a lot of sense. Yeah. Well, I wanted to ask you about some other fun topics and not just focus on brain discovery. I saw an article you wrote a couple of years ago. We were celebrating getting to a million views on LinkedIn, and you were sharing some tips about, you know, some things that you had done that kind of helped you along the way. If you rewrote that article today, what are some things that you would add to it.
Cat Casey:
You know, I think I mentioned it. But I think the single most important thing is to forget about views. Forget about the network. Foster conversation. I'd rather have 10 or 15 really great comments, or have like, I have 30,000 followers, but probably 1000 of them are core people that I interact with on a frequent basis. Foster that and. Network, because I go to that network for advice, for input, for insight, like my why was, you know, I was 23 when I started using LinkedIn, and I was nervous to talk to all these people, but on LinkedIn, there was no barrier. I could add a comment, ask a question. So use LinkedIn, which, you know, not for virality, because that's up to an algorithm that they change every six months, but to foster a really good network of people that you're actually having a conversation with. So occasionally I'll catch myself. I'm like, I have told people a lot of stuff in the last month. I haven't asked any questions. Shame on me. So, you know, ask questions genuinely interact with people, elevate other people, and then find who you is, authentically you like I I'm kind of known for, like the bright colors and, you know, a certain way of talking, and I'm very consistent with that, because it's who I am. And by authentically being who I am, if people meet me in person, they're not surprised. But also, there's a familiarity, like we have a relationship, a digital relationship, that doesn't feel fake or, you know, like a lunatic of LinkedIn. I follow them on Reddit. It's my favorite subreddit, people that are just self promoting, saying, Oh, I'm so humble and so proud all the time. Don't do any of that stuff. Just be honest, be candid. Share your wins, share your losses, but also ask questions. Ask how you can help people. I think you know, my network has grown pretty substantially, probably 4,000% or something crazy like that, in the last five years. And it was when I went from I'm going to present a polished image and only share articles and only say things that HR won't get mad at me about to I'm just gonna be honest. I'm gonna be candid. And, you know, most people don't think of lawyers and hot pink and sparkles. I don't care. That's sort of my jam. Figure out what your jam is, and it may not be outgoing, it may not be being an extrovert, it might be a little bit more professorial. But if that's you, lean into it and you'll find the people that are drawn to you when you're authentically yourself are more valuable for you as a human for your career development and for, ideally, what you're doing with LinkedIn, which is using it to kind of create a digital network of friends, colleagues, peers and mentors.
Ryan Bankston:
Yeah, that's awesome. That's awesome. I like that. I like that answer a lot. That engagement, like creating, like, a the inner circle or whatever of your of your followers that you're engaging with commenting on their posts, they're commenting on yours, and you're building that trust and relationship with them is I've seen that just in a little time that I've been on LinkedIn, and I'm tiny, you know? But, but like, if I put up a post and it's worth a darn, I can pretty much guarantee, like, Colin Levy's gonna make a comment, you know, yeah, same as I see him. It's like, so it's, I know, like, there's a lot of these legal tech folks. It's like, very generous, you know, and I interact on LinkedIn. It's like, we get something more out of it than just like, Oh, look at me. Look at me.
Cat Casey:
Well, you know, I like to joke. I'm kind of a disco ball. Like, there might be a spotlight on me, but my hope is, if there's a spotlight on me, I'm then shining it on people in 20 directions. And the other thing is, like, the algorithm changes a lot, but it's not complicated. Like, I know how I could go viral every single day, but the viral posts aren't helpful in my community, and they're not really helpful to me. Like, take, for example, last year, my number one post was my foot, right? I won't even go into details on the whole foot, whatever I literally, I had torn my Achilles tendon, and I was using it like a laser, and sort of my foot was in a laser. And I was talking about, like aI generated, like inorganic stencil. That was really cool. But then I looked at him, like, I have 80,000. I think it was ultimately 120,000 people that looked at this post, because there's this post, because there's a foot in it. That's not helpful to anyone. It was a cool thing. And so I could have done a whole series about five more pictures of my foot with a laser and had a million extra views, but that's not helpful. And so, like, I kind of try to balance the you know, my number one mission today is this, is this anti Renaissance is really scary for lawyers. We've never been or legal professionals, not just lawyers for the legal industry. We've never been the ones displaced by tech advancement. And so there are millions of people that are scared. And my entire mission is, I want to talk about the big scary things and break them down in a simple way so they're approachable, and people feel like, okay, at least I understand this, and my foot does not further that goal. So, like, that's kind of another thing you kind of want to keep there. Like, sometimes it's easy to fall into, oh, that posted really well, I should do more like it. But really, like, ask yourself, is this talking to the people I want to talk to? Is this helping people? Is this helping me? And if not, you might have to say, bye, bye foot post.
Ryan Bankston:
Yeah, that's awesome. Like, I mean, you've got to keep yourself happy too, right? Be proud of what you're putting out there, whether it's a foot or not.
Cat Casey:
I mean, it was a cold post. I just didn't need to do any more of them. They don't regret it.
Ryan Bankston:
It wasn't about the vanity metric.
Cat Casey:
I'm like, There's a loser, there's them, and there's my foot, awesome. But then I was like, when I dug deeper, I'm like, Oh, I can see why this, you know, hit the algorithm. It's not, it's not doing what I want, which is educating, you know, my universe of people. So hey, I'm gonna pivot, but that's my other thing. So like, figure out who you are, figure out what you want to do, have a conversation, and then even if something does really well, if it's not fostering that, then, you know, say goodbye to the likes and maybe do something different.
Ryan Bankston:
I like some of what folks have talked about with, like those vanity metrics right about and being too focused on them, because, like, what, at the end of the day, what are you really here for? Is it to help somebody? Is it to get some business? Is it both, or do you just want the attention? Maybe that vanity metric is what you're here for. There's people out there that I think that is what they're here for. I had a great meeting with some other podcasters at the beginning of the year, and we were talking about promoting our podcast, and that became a part of the conversation. Like, what's the number of people that look at the podcast and watch it, download it, whatever? Is that, like the end goal, or is it the one person that listens and actually gets a takeaway from it? And then, like, how do you actually measure that you know? And sometimes there's no way to measure it other than your own intuition. So I like your advice. Like, if you're putting something out there and you don't feel right about it.
Cat Casey:
Well it's also it's like, it's really having a clear grasp of your why. Like for me, my Why is, like, I'm a very good translator between tech and legal, and so I want to use that to help people right. And so that for me is my why. Like for you, if it's Hey, I think this podcast will help people looking at the vanity metric of, is this going broader? Is the way I'm communicating about it increasing over time? That's great. Like, it's about balancement. Like, I think it all comes back to, am I having a conversation? Am I doing my goal, whatever that is? Am I being authentic? And if it's those three things, then you know, you will see growth as a result of that. So it's not that the views don't matter. More people are seeing my stuff today than they were when I first wrote that during COVID. But it was by recalibrating that I made sure it was growth in the areas I wanted to see growth and reaching people I could be helpful for. Yeah. So you know, growth is not a four letter word, both literally and before.
Ryan Bankston:
Oh gosh. Well, you've been doing so well. I am so intimidated by your travel schedule and all these conferences you do. It's like, I just don't know how you get, like, work done. I can imagine sometimes it's on the plane, on the plane, on a break from Sir conference, like, running back to the hotel room for something. But how many conferences did you do last year?
Cat Casey:
So I think between, like, virtual and IRL, it was, I think I averaged like one a week. I was, I know it slowed to go around the globe three times. It's kind of lumpy though, like, I'll have, I had a couple weeks where I hit like three events in a week, and then there's a couple weeks in the summertime where it's kind of quiet before the big ilta push. We're about to kind of really, we're beginning the crazy conference season. Well, we'll be in full swing by the time this airs, I think legal week is next week, but we've got to evolve. I think the University of Florida just happened like we're in the beginning of the first half of conferences. And then there's a second one. So yeah, it's constantly on the go, but it's, I also get FOMO. If I'm not at an event, I'm like, oh, I should be there. If you're talking with people, I should be helping them out. And so it's, I always say I'll travel less, and then I'm like, or I could still go to that other conference.
Ryan Bankston:
Wow. Well, I wanted to ask you, from your experience with these conferences, which one's the most accessible and valuable for such a low and small, firm return.
Cat Casey:
So it depends what they're looking for. If they're looking for some basic education and kind of connecting with people, I recommend small regional whether it's Association based or the smaller niche ones like evolves a good niche when it's not that small, but it's kind of a little bit narrow, where you can actually connect with people, but if you're looking for a big platform to access the most people, like the main ilta, not sponsoring necessarily, but maybe sponsoring, but just attending, you're more likely to meet a broad cross section of people. So I think balance it between the low cost regional either you know, you. ABA or WBA or masters, or any of those smaller regional ones that might kind of be, you know, when they discover whatever a niche within your geographic area. But then, if you're looking to kind of connect, understand the tools and technology and the big trends, and you want to level up real quickly, that's where, you know, the iltas of the world can be incredibly powerful, and your podcast focuses on CIOs. It's my number one recommendation for CIOs, I think for practitioners that are very e discovery focused. There's a ton of value in legal week, which is again happening next week. But for CIOs, I know when I was at Gibson, it was the number one event that my CEO for a bigger firm, but he always wanted to go do, because you could see the macro trends of what tools and technology can help you, kind of you know as your force multiplier, and also have an access to a broader swath of people in different geography, sizes of firms and such. So balance the two, and there's a ton of free stuff too, the clees, the webinars, use that because some very, very brilliant people are literally 6-12, inches away from you through the screen at a lot of the virtual conferences and virtual events you can attend.
Ryan Bankston:
Yeah, absolutely. That's great advice. I've been a member of ILTA Since 2012, definitely a big fan of the community and the conference when I've been fortunate enough to be able to go to it. One thing I noticed there that was actually pointed out to me in 2023 when I was in a session about the technology report they do every year. A lot of the small firm groups don't go to ILTA. You know, it's, it's, well, you start seeing Clio there now, but they haven't for years. They haven't gotten wet , you don't see smoke balls there. Yeah, typically it's bigger like net documents and I manage, you know, or after.
Cat Casey:
What I'll say is, they're there. They just may not be sponsoring, sponsorships are kind of expensive. So if, if you have a smaller legal technology company you work with, not that Clio small, there probably is a team there. They're probably some event of some kind. What's nice for the smaller firms is that you might be able to have 10 or 15 meetings, even if it's not a, you know, our key sponsor. But also balance it out like it is, there's a cost to it. It's not an inexpensive event. It's super valuable, depending on what you're looking for. But there's also regional, niche events. There's, you know, there's like the GC Forum, which I think is San Fran. There's a bunch of individual city and state a sense groups, Association of Certified discovery specialists, women, you discover ALA, which is Association of Legal Assistance. But it's also like paralegals. So it just depends what you want, kind of, look, what's in your geography. And if you've got a budget, which is great if you don't, there's a whole host of regional stuff that you may even have to travel out of town for. And then virtual like, I think with COVID, the quality of virtual content skyrocketed. There's a lot of inexpensive free CLE there's some pay to play, ones like Ali and PLI, Aced, EDRM women, you discover I'll all do different stuff. And then there's even virtual or hybrid conferences, like the University of Florida conference where actually 1000 people attend virtually. 200 attend in person. So there's a mix of stuff out there, and there's no one right way to Tetris it together. It's just about getting up and connecting and connecting and learning.
Ryan Bankston:
Well which one of the conferences are your personal favorites and which are the most productive to reveal?
Cat Casey:
So my absolute favorite is legal week, which we're heading into. But I may be biased, because they literally have built me a personal techno cat stage, and I just get to have, like, wordy podcasts, and we kind of, that's the reveal takeover. So where will I be introducing or either me or Wendy will be introducing Rob Lowe for the keynote. So we're very dialed in there. So I love that. And then I spent a decade in New York, so it's kind of a homecoming. But that one I see is really good for big macro connections, and it's one that is expanding into contract analytics. It's a little bit more eDiscovery focused. Another great one, if you're a little more senior, you're looking to kind of just brainstorm, is there's the Sedona conference, and then there's EDI, the eDiscovery Institute, both great University of Florida's ediscovery conference, again, when I think it'll have just happened really, really great, a lot of brilliant people regionally. There's the Masters Conference, which is legal, tech, cyber and E discovery. It hits 12 cities throughout the year. Solid is great. It's a little bit more aimed at corporate and kind of is cross functional across different tools and technologies. And it does TED Talks. So seven minutes, go, go, go. And then my absolute favorite is a new one that just started that is actually called Women in AI, you could come as an ally, but it's sort of focused not just on AI and legal, but different ways women in AI are across finance and medical and legal, kind of reshaping the industry. So that one was a great one. All shout outs to the other cat. Cat moon. We occasionally get confused. It's not like the Highlander. There can be more than. One, and let me think, what else? Oh, and, of course, Ilsa, obviously your shirt reminded me of ILTA and ILTA evolved, are great, and we've worked with them forever. So there's, there's this whole universe. And, you know, net diligence. There's all sorts of, I'm thinking from the lens mainly of like, I've 20 years of ediscovery. So those are the ones I think of. But there's ABA Tech Show, which is a little more broad. There's a whole universe of contract, analytic and cyber and compliance focused ones, and there's no one wrong way to go. They're all pretty good.
Ryan Bankston:
It's a lot of high praise in a lot of conferences.
Cat Casey:
You know, it's funny, like, I think I tried to pinpoint every conference I went to, and they all kind of blurred together. I couldn't even just search by like, con like, I had 75,000 pictures, which was a problem in and of itself, and I couldn't even search to make sure which one was a conference day versus not.
Ryan Bankston:
Wow. Well, what other social networks other than LinkedIn are good for you, for business?
Cat Casey:
Yeah. So my main jam is LinkedIn. I sort of am good at that. I have played around a little bit with tick tock, so initially I pretty much used it because it was easier to edit videos than any other app. Application. There are, of course, concerns and considerations about it potentially being a vector for Chinese misinformation, and the Chinese state owns it, so be careful with that one. But there are a lot of talk communities that's pretty, pretty active. There people like Alex Sue post often. X Twitter, the artist formerly known as Twitter. I've never really figured out how to do it very well, so I'm not a big proponent of it. I've got maybe 1000 followers, and I tweet very occasionally. But folks like you mentioned Colin Levy. He's very active there. And there's a pretty good legal and legal tech community. I don't have Facebook, I have Instagram. I even have a techno cat specific Instagram, but I don't really use them as much. It's not as active a community. I've played around with some of the different discord servers, which, again, that's a little bit even dorky for me. They're not as active. And I'm sure, like we tried when what was it? What was the audio only? I tried one that was not great. Where it was clubhouse. I think that was last year, the year before. But honestly, I feel like the community, it's strongest on LinkedIn, and then also LinkedIn groups like, I've got an AI baddies group, which is, you know, focused on bringing out people that, like AI and AI Esquire is another group I have where it's just, how do I as a lawyer embrace technology and keep myself in the loop and in the driver's seat? So there's also subgroups within LinkedIn that are really good for just kind of sharing more tailored information. So if anyone wants to join them, just ping me. But yeah, I feel like for us, LinkedIn does a really good job. It can be annoying, it can be fun, but it's a nice community that's been fostered there.
Ryan Bankston:
I'm most comfortable on LinkedIn, definitely. I mean, it's, it's, it's very clearly, like, designed for work, it's a professional network, so it's like, that's what you're in it for. Why not just that's the place to be, you know so well. And I've always felt a little awkward, like trying to promote business on Facebook, you know? So I'm like, oh, that's where my high school friends are. Family is. And it's like, you know, posting about politics, and I don't want to mix that.
Cat Casey:
I think, like for me, I noticed a long time ago, a lot of my best friends were similar to me. They were in the same industry, similar interest, and so we have kind of that I'm okay bringing a little of the personal into the professional. I don't like you were saying. I don't like bringing the professional into the personal. That's why, for years, I actually had an ediscovery cat, because that was my old name before I translated it Facebook, like as a decoy if a client asked me if they could connect on Facebook, which, shockingly, they did 10 years ago. So I really separated church and state, but with LinkedIn, it's like, as I've kind of been more authentic and candid, I bring a little bit of the personal, not too much. It's not Facebook, but I feel more comfortable going that direction than the other. It's a balancing act. With Tiktok, I literally only post videos I'm going to then reshare on LinkedIn. So I don't know if I'm tick talking, right.
Ryan Bankston:
I get in trouble for not, for not using it. So I always have somebody else send me a link, and sometimes I can play it, sometimes I don't, because I refuse to install the app. And they're like, why not? Why don't you do it's always this big thing.
Cat Casey:
My caution is, the algorithm is probably the best algorithm out there for social media for dopamine response. Like, I will go on to get a few ideas and blanket it'll have been an hour. Like, it must be doing something with my brain where it's just incur, like, it's the positive reinforcement it is giving. Is very potent. LinkedIn, I can do it off.
Ryan Bankston:
See, that's why I don't do it. I know myself, I would sit there and doom scroll for two hours. Like, you know, I've got to, I've got stuff to do. Same, same, awesome. Well, thank you. This has been such a great conversation. There was something else I had written down I wanted to ask you about. I overheard a big firm attorney talking about limiting the other side's ability. Use AI tools like litigation stipulation, like, but that's not something I'm super familiar with, so I wanted to ask you about it, like, how would that work? Well, I don't know, just in your opinion. Like, how would that work if you're on the team and suddenly you're limited because of the other side stipulation?
Cat Casey:
I think it'll be really challenging to do, because there's such a delta in the understanding of what even AI is. We've noticed it even a step back, not even from opposing parties making stipulations, but from judges making rulings about AI disclosures. They vary from disclose, if you're doing research using generative AI, to tell me every time you use AI, which, on its face, doesn't sound too crazy, until you're like, well, the GPS I used to get here is AI, the Google search. I use the Lexus search, I use the spell check, I use, that's all AI. So I think there'll be challenges for opposing sides to effectively tailor what their stipulations are. And I think as Gen AI and AI in particular becomes more integrated into legacy tools. Like, it's available on Lexus and Westlaw now it's getting baked into, I want to say that documents have functionality a lot like, I mean, obviously purview as well, and co-pilot, a lot of basic business tools have Gen AI built in. It's going to be very hard to try to exclude that. I think you'll see people negotiating ESI protocols on if and how specific functions are used, like, you know, we explicitly agree or do not agree, to have any sort of Gen AI, automated first pass review, that sort of stuff. I can see having teeth, but opposing sides, saying you can't use AI is going to be very challenging, because, one, we're already using AI extensively. People are confusing generative AI, which is AI that makes stuff and writes stuff and creates stuff with AI, which is AI that connects the dots or makes the recommendation. And so because of that confusion, they're thinking, Oh, it's only new things they're going to use, not realizing all of these tools are very stable. Standard Business already uses AI extensively. So if you say don't use AI, it would be like saying don't use the internet, right? It's very challenging. Or, you know, you can't use the cloud, right? It's a little too broad. So I think what you'll see is courts pushing back on over broad stipulations or specific language, say, in an ESI protocol, because the people making it don't understand what AI means in legal and legal tech, or in business two, I think there's an evolution of the acceptance of AI. I always joke, like, three months after chat PT was released, there was a South Park episode written about and written by chat GPT. And you know what happened after that? My nephew, my mom, the lady at the United airport lounge, started talking to me about AI. It's gone mainstream, and it's hard. It would be like telling someone in 2004 you can't email anything. You can't research anything online. The ship may have already sailed, right? We already may be too familiar with and comfortable with using AI. So what I think you're going to see is those broad stipulations will probably be shot down. Even judge disclosure requirements of all AI shot down, but you might see specific language about you can't use AI in this specific manner. You have to disclose if you're doing it in this specific way, and or you have to validate that there was a human in the loop, a human in the driver's seat, a human queue seeing that it wasn't like the Avianca case where the guy copied what chat GPT said, pasted it and submitted his motion. That wasn't bad. AI going rogue. That was bad. Lawyer being lazy.
Ryan Bankston:
That case is talked about pretty often.
Cat Casey:
Well, you know what I keep saying, people missed the boat. Here's, here's what the guy did. His son said, chat. PT, is awesome. How did he validate that? He watched one YouTube video, and then he took what chat P said, pasted it into emotion. And then when the judge called him out for it, he asked chat PT, was the citation Correct? To me, that's a lot of laziness, because, you know, you can ask, also ask to chat GPT, please provide me with the link to the case you're citing. Click the link and read the case, which is what you would do if a first year associate made you a motion or a brief that sounded too good to be true with cases you hadn't heard of. So it's, you know, people are afraid that AI is going rogue. That's not an AI going road. That's lazy lawyering. We've had that forever.
Ryan Bankston:
Yeah, you mentioned disclosure earlier, and that's something that's certainly talked about a lot in some of my circles down here. I want to ask what you think, but then also, like, what came out of that task force, as far as, like, when should an attorney or practice disclose? And like to who their clients are.
Cat Casey:
I focused more on the risks and opportunities. That was the section that I was the kind of the co chair on, or the chair on, actually disclosure was it was talked about a little bit, but not quite as much. It was more like a prescription of use, do not use, but looking at the. Disclosure discussion, I think what it comes down to is an AI being used in lieu of human legal decision making. Is there an audit trail and transparency? Is there a way for the output of the AI to be validated and you have to disclose if there's not humans in the process now, using AI to solve the blank page problems? You know, for past at the wall, get some ideas, probably don't need to disclose that. Using AI for final case citations probably do need to disclose that. So I think what we're going to see is, right now, there's a lot of very broad disclosure of everything that is not going to stand because, again, spell check uses AI, your spam filter.
Ryan Bankston:
You're even using like in like reveals Advanced Learning Tools, stuff like that for calling like that would.
Cat Casey:
Email threading is AI, right? Or Boolean search in LexisNexis AI, right? All of that's AI. So if a lawyer could actually say I have nothing to disclose, it would be like malpractice, right? So the over broad ones are not going to be there where I think the rubber is going to hit the road. Is the use case, the transparency, whether there's a human in between the AI and the output, and clearly disclosing, you know, when it's a kind of an alternative workload. So if you're using, say, Gen AI for automated first pass review, that is different from, hey, I was using the Gen AI search function to summarize cases before I read them right with Lexus or Westlaw. So it's, I think we're going to have to get a lot more granular and these broad kind of disclosure rulings, or, you know, proactive, you know, disclosing to the court, they're going to have to get a little bit more nuanced, because AI is such a broad universe, I'd be like saying disclose if you use the internet, you know, it's too broad, you know.
Ryan Bankston:
So when it comes to Westlaw and Nexus. Which do you prefer?
Cat Casey:
You know, I am going to plead neutral on that, because too many people like both sides, period. I plead the fifth there. But both are doing.
Ryan Bankston:
Do you think it's a good idea for any firm to sign a three year agreement with one or the other in this day and age where everything is changing so much like, what if they sign a three year with Lexus and then they're like, but co counsel is great.
Cat Casey:
You know, I think licenses make sense from a co innovation standpoint, right? So these big players want to be dominant, like, it's not like Lexus will innovate and Westlaw will not. That's not a scenario. So you're going to want to have a relationship where you can help build the tool that works better for you, be part of you know, building solutions that work better for you, waiting for seeing who's going to win means you lose, because then you're behind the eight ball, and if you make the wrong decision, like you could look at renegotiating or I just don't think, I think it's like if someone were To look at using open, AI versus Google, llama versus Claude. Those are foundational AI models. They're a little bit different. Some are better than others. Each new release is a little better than others, but they all have the same fire under their metaphorical ones and zero. But to innovate. They're all innovating. So whichever one you pick, you're going to be ahead of not picking one to develop with. So I don't think there's a negative to picking, picking a side. I guess I'm just not gonna do it.
Ryan Bankston:
I hear you. Well, this has been awesome. This has been such a great conversation. Thank you so much for joining me today. You're so generous with your time and generous in this community. I certainly encourage anybody that's into legal technology, whether you're an attorney or not, follow cat Casey. Follow cat Moon too. If you want both. The cats have some great material. How would they find you if they wanted to find you, Cat.
Cat Casey:
So I am Kat Casey on LinkedIn. If the background is hot pink, that is definitely me. I also got a website, www, dot the technocat.com, and I am the techno cat. I think it's the underscore technocat on Tiktok. But LinkedIn is usually the best way to find me. And I'm pretty liberal with adding people, as long as you don't ask me to add you in Cyrillic or in Mandarin and I can't read it, then I'm going to probably add you.
Ryan Bankston:
That's wonderful. Thank you so much for joining me, and thank you for listening in today, some listeners, I hope you'll find me again next month. I put out episodes every month. This is Ryan with CIO.legal, and if you think that your firm could benefit from a part time CIO, look me up. I started 200 a month, month to month. You can find me at CIO.legal. Thank you.