Is AI Drafting My Arbitration Award?

[MUSIC] Welcome to Resolutions, the official podcast of the American Bar Association, section of dispute resolution. Every episode is a journey into the heart of conflict resolution guided by our diverse community of experts. We're here to explore innovative strategies, share wisdom from the field, and inspire excellence in resolving disputes. Whether you're a dispute resolution professional, a curious learner, or someone passionate about fostering understanding and peace, you're in the right place. Let's dive into today's conversation and discover how we can transform conflict into opportunity together. I'm your host, Darren Gottholf. This week I'm sitting with Clear Brief AI Founder and CEO, Jackie Schaefer, to discuss is AI drafting my arbitration award. Jackie is the Founder and CEO of Clear Brief AI, a legal tech startup helping legal writers draft credible documents with hyperlink citations in word. Clear Brief AI was named the 2023 Legal Week litigation product of the year, and Jackie was recognized this week by the ABA in its 2025 class of legal rebels transforming the justice system. Clear Brief AI is also a finalist in the 2025 Legal Week Awards for the best use of AI and e-discovery and litigation. Jackie is a finalist for the 2025 Innovator of the Year award as well. Wow, Jackie, that's a lot of awards. Welcome, Jackie. Thank you. It's a word season in legal tech. Yeah, just full disclosure before we start, Clear Brief AI has recently partnered with the American Arbitration Association to provide its services to AAA arbitrators. While today's discussion is not going to be a sales pitch for Clear Brief AI, I'd like Jackie to briefly explain what our product is and how AAA arbitrators will be using it if they choose to do so. Absolutely. Thank you so much for having me here today. This is wonderful. Clear Brief is a Microsoft word add-in, and at a high level, the goal is to keep the evidence visible during every stage of the writing process. So Clear Brief is unique in that it's been specifically designed for litigation, for arbitration, so that it sits with you inside Microsoft word, and it's used from the blank page when you're importing large massive volumes of documents all the way through the drafting to site checking. You can use, if you like, certain curated generative AI features, but we have a lot of classic AI tools that are not generative AI that focus on things like site checking, creating exhibits, finding other evidence that could support your ideas across all of your uploaded documents. And then at the end, the user can share a hyperlinked version of their pleading with the arbitrator. So that is a very exciting improvement, honestly, to the world of litigation, to arbitration, and the arbitrators and judges absolutely love getting hyperlinked pleading, because as you know, in an age of AI hallucinations and people potentially citing cases that don't exist or citing to things that do exist, but for the wrong reasons, being able to hyperlink every single reference also to the facts of your case really makes that that piece of writing so much more credible and trustworthy. And I believe that was the famous case at a New York where some attorneys, I think, on the fly were using one of the AI sites and it hallucinated and generated a bunch of fake cases that didn't exist and they got caught. That's obviously why hyperlinking citations to real cases is really critical when you're drafting your briefs. Yeah, and honestly, I feel like there's a new announcement pretty much every day, every week, in 2025, we've had some really ironic ones with a Stanford professor submitting an expert declaration that referenced nonexistent sources in a case about AI. We've had even just today and other news about a major law firm being sanctioned for fake cases. So it's really ubiquitous these days. And honestly, what I have been trying to explain to folks is that one of the things that classic AI tools, so natural language processing versus generative AI that's sort of writing text, one of the things that classic AI is really good at is catching mistakes like this. And so that's where I work in this new era where you need AI to help you avoid AI hallucinations, right? For example, clear brief will automatically detect, oh, this is trying to cite to a case. We'll cross our reference, we partner with Lexus Nexus and Fastcase VLec. So we'll look through all these case law repositories. And if it's a real case, it will display automatically. If it's not a real case, it will say, we can't find this. You might want to double check this. So there's some interesting dynamics now where AI can actually help us avoid embarrassment caused by the generative AI. And we'll get to it later in the discussion. But yeah, I think this also goes to part of the discussion of using free tools versus some of the paid tools and kind of the difference in the results that you might get. All right. So let's start with the basics. How our attorney is using AI today? Sure. There is such a wide variety right now of how folks are using it. I still we haven't quite reached full adoption across the industry yet. But what I'll say is that I founded clear brief in 2020 before this latest AI hype cycle. And we've never seen adoption like we're seeing now. So I do think that 2025 everyone realizing this is not a flash in the pan. This is something we have to actually take seriously. Our firm needs to start having an AI strategy. And we focus on litigation. So what I know in terms of how our customers are using it on a daily basis. It's for drafting the fact section of briefs. It's for it's being used for summarizing deposition transcripts, taking 500,000 documents. And you can press a button and get a timeline with summaries of everything that happened on each day. It's all organized for you with links back to the source so you can easily see it. Things like that really help you organize all of the massive amounts of data that we're getting. They're also using it like I said for site checking. And we are also seeing in the transactional world. It's being used for contract drafting due diligence. Really most of the tasks that we're doing that are very repetitive tedious work. There are AI products now that are being built to to focus on those workflows. And we're also starting to see it with the legal research and e-discovery. And there are a number of products that have come out that are essentially we all remember in the old days you were a first through third year associate law firm. And what's one of the things you have junior associates do. They sit in a room with a bunch of other junior associates and they basically do doc review, right? And that's paying humans many hours to basically read things. And with some of the AI technology that's coming out it can very quickly summarize a vast amount of data that's fed into it and defined hopefully what you're looking for. Also language translation, right? A lot of times you need you have legal documents that need to be translated from one language to another. AI has the ability to do that. And even just you mentioned transcription, right? I don't know if we're going to completely put court reporters out of business, but traditionally that was a position where you had to have a person with a keyboard typing along with your deposition or in the courtroom. And we're seeing AI going into that space. All right, I know law firms and attorneys tend to be on the cutting edge when it comes to adopting new technology like this. And sometimes the courts can be a little bit slower to catch on. Are court systems using AI and how are you seeing it being used? Yes, so I do know that they are using it. We have a number of court systems that use clear brief. And I would say the workflows that courts are focused on are more around, they're quite similar to how the law firms use it, but they are really focused on checking over their work product, organizing the facts to quickly find evidence to make sure that they've adequately supported their conclusions in their opinions and even when they're going to deliver an oral ruling, they want to make sure that it's very accurate. So courts are really focused on using AI for search, for accuracy, for checking over pleadings and citations. And to also organize the vast amounts of evidence that are being filed in these cases, because it's everybody is struggling with the volume of discovery that a typical litigation matter requires these days. Now a lot of court systems probably vary in the levels of technology that they have or that they use. Certain states fund their court system is better than others. Do you know of any courts adopting the technology faster than others that you'd want to highlight? I'm not able to speak about that unfortunately, specifically, but what I will say is that I am seeing different working groups, whether it's part of nonprofit organizations that courts are a part of where they are forming working groups to experiment. So I think everyone in the courts are definitely in an exploratory phase right now. I'll say I did have the opportunity to speak at the National Conference for Chief Justices of State of Public Courts. And that was a wonderful opportunity to talk about AI, to talk about real-world use cases. And I would say there's definitely more openness to using AI than before, but courts have a serious budget problem. They have a serious bandwidth problem with all of the responsibilities that they have already and just the technical debt that they have and that they're trying to catch up with basic technology that they need to install for day-to-day operations. I do think it's a bit of a slower journey. Well, so obviously to tie it into resolutions and resolutions is a podcast for the American Bar Association dispute resolution section. I'd like to ask you how are ADR professionals using AI? Absolutely. So I see ADR professionals using it beforehand to review the party's pre-mediation pleadings, their briefs, and to help evaluate the strength of the evidence and the arguments before they get to the mediations that they're well prepared. And then having the ability to search through facts very quickly is really valuable as well when you're live inside a mediation and being able to point one of the parties to an important document that might help get them thinking about other possibilities or helping them view another perspective that can be really valuable. And then to the extent that they are writing up findings, it's also very helpful. Now, one of the reasons people choose arbitration over traditional litigation is arbitration is supposed to come with a level of confidentiality that's not provided by the general court system. And so I guess my next question is AI safe when an arbitrator is running information that he or she has received from the parties through some sort of AI system. So it will always depend on what technology they are using. So this is a new era where any professional who is using AI and illegal profession needs to start understanding better the terms of service of the tools that they're using needed to understand the privacy implications of using those tools. I am still surprised sometimes to hear about lawyers using just open open tools that are available for free online. Those are usually not secure because they say they'll be able to use your data for training large language models and things like that. And those are some of the things that legal professionals are now getting more attuned to before they start using the data for legal work. But for example, clear brief, we are sock to type two certified. We meet the strictive security requirements. We're already used, like I said, by big law, by courts, by government agencies. So we've had to we've been put through the ringer with our architecture and the way it's designed for security and confidentiality. So then that just goes to fits nicely into my next point. We have this saying, right, if you don't pay for the product, you are the product. So folks that are using a lot of the free products, it is using a lot of the information you put into it to essentially learn. And if you have concerns about data security and privacy and with cases that you're handling, you're going to have to probably think twice. And as always, read the the service terms and conditions of whatever website you're working with where they're supposed to explain essentially what they do with the data. So it shouldn't be a surprise to you. So in response to all this, in 2023, the American Arbitration Association released the principles support and the use of AI and alternative speed resolution. Now, I'd like to briefly mention each of them for your comment. And I'm going to paraphrase obviously to cut down a length. I don't want to just be reading a document the entire time here in our conversation. So the first one is competence and its competence in integrating AI into ADR frameworks requires legal professionals, arbitrators, and mediators to ensure they are proficient with AI technologies and understand the risks, benefits, usages, and ethical considerations. Anything to add to that or comment? No. And what I'd say is that I hope folks sometimes who are nervous about emerging technology, when they hear about any of the risks, like what we just talked about, they think, okay, I'm not going to touch AI. And so I don't think that's the right approach. It is really important to educate yourself, because maybe not even just about getting your work done more efficiently. But a lot of the cases that you might be evaluating have legal issues related to AI to these technologies. It's just an area that we all need to become more familiar with. So I also even recommend using some of that we talked about, the free kind of general-purpose AI tools like ChatGbT. I'm not saying don't use those. Those can be really helpful for just experimenting with and learning how tools work. Just you can use them to do things like summarize this email from my kids music teacher. But there's all different non-legal use cases you can use to play around with those tools. It really will help you get more fluency and understanding how they work, which can have benefits for your legal practice. Yeah, and I was going to say one of the things that concerns me the most about people in the ADR field using AI is using it not necessarily knowing how to use it. So as you mentioned, maybe practicing with information that isn't so sensitive on one of the free sites and see how it goes before you start actually taking information from your cases and feeding it through various AI tools. Okay, the next one is confidentiality. Safeguarding sensitive data is critical in ADR as in business in the law. Yeah, I think this one goes without saying that everything that you do, you need to make sure that you are up to date and using typical and modern security protections on your computer, on your systems and the products that you use. And one of the things that is also important, for example, with clear brief, we do have a customer success team. We have real people who can hop on a call and help you and walk you through how to use the tool on real documents. Everyone is under NDA at our company. So to know that if you are talking with folks at a vendor, make sure that they are under NDA and everything remains strictly confidential. Excellent advice. Okay, the next one is advocacy. When using AI tools an advocate should promote accuracy, expedience and candor, align AI applications with the best interests of clients in the justice systems integrity and champion responsible AI use as an essential component of advocacy. Yeah, I think this makes a ton of sense. I completely agree. I think when we already have talked a little bit about how AI tools can actually support accuracy, that's one thing that computers are great at is finding inconsistencies, pointing out mistakes and those sort of things. I think this value is really speaking to the idea that it isn't just a fringe concept anymore to use AI to use technology. It's actually an essential part of your day-to-day work. The next one is impartiality. ADR Neutrals must scrutinize AI data impartially, avoiding over-reliance on technology or outputs provided by one party or one vendor. And I was thinking about this the other day. But you can take the same prompt and feed it into chat, GPT, Claude, Gemini, whatever program that you're using. And you can literally get three very different responses. If you're using this technology, let's say for legal research, it might make sense to use more than one type of legal technology. But if there's anything you want to add there. Sure. I think one of the things that is helpful to think about is product design and whether the design of the tools you're using are supporting your ability to scrutinize that output and be impartial about evaluating it versus just simply making it by the way the information is presented, making it seem like it's authoritative. I'll just give kind of some examples about Clear Brief that there are times, for example, if you select any text in a document you're drafting, Clear Brief's AI can find other places across all of your documents that seem to relate to or could prove that idea. When it's bringing up those pages, it ranks them in the order that seem most helpful and it's very clearly listed as suggestions. These are just suggestions and only you can really know which piece of evidence is going to be most helpful to support your idea. We also have color coded scoring that it's one of our patented features that flags how well a certain source seems to support what you've written. And we put a lot of thought into that color coding where it's not designed to, for example, we don't use red, bright red because that's too alarming for people that there's going to be judgment about why you might cite to a particular documentary page. So that's just how I've been thinking about it as well. That product design should leave room for the human judgment. And then that relates to independence independence is our next one. AI can provide valuable insights. ADR professionals must exercise independent judgment. And this is something that I always remind folks. And I think it's a fear within the legal community, especially with attorneys when arbitrators might be using some sort of AI technology on the case. The AI is not supposed to decide who wins and loses the case. And you're not supposed to be asking AI to rule on your various motions along the way. You still have to use your own independent judgment. I don't have anything to that. I don't have anything to add. I think that says it really well. Okay. All right. The last one is process improvement. Embrace AI is a tool for enhancing accessibility, efficiency, and fairness of ADR. Yeah. Absolutely. I think that's one of the overlooked aspects in a way of AI that as lawyers were so ready to jump on all of the issues and concerns people have with it. But there are a lot of opportunities in terms of improving the process. One of the ones that I mentioned at the beginning is hyperlinked pleadings. Why aren't we all using hyperlinked pleadings? So it's easy for everybody to see exactly what we got our information. It's why don't we haven't we done it always because it was very tedious and there wasn't an easy way to do that. But with the advent of technology, you can do it right inside word. You can every single factual or legal source can be made clickable so that you can easily understand the origin. And that's a really helpful process improvement. When we were talking about translation services, that's something where it can improve the fairness to have, for example, I'm on the Washington State Bar AI committee and one of the things that we've been talking about and thinking about is there's so many different languages including languages that it's very rare to and extremely expensive to be able to find translators for some of the languages. And the workarounds are maybe not as helpful for the litigants. So can AI help us get incredible translation for a language that's more rare in to have translators available? That would be a wonderful process improvement. So there's a lot of opportunities that we should be thinking about. And yeah, and that's also we're going to discuss just a bit access to justice and all that. But that's that goes along with an access to justice, access to justice issues. A lot of people in this country, English might not necessarily be their first language. We have a lot of prosay claimants that are handling their own cases. For example, at AAA and English might not be their first language and all of a sudden they have to start drafting legal documents, which is something that is hard enough when you're not an attorney. Can you imagine having to do it in a language that you don't speak is your first language? So that's there's an area there where it's definitely very helpful. What excites you about AI? Sure. So having been now a startup founder for the last almost five years now, excitement honestly still hasn't worn off of the thrill of showing a legal professional a better and faster way to do something that was stressing them out. It was going to take them hours. It is so rewarding to be in the room with someone even virtually. It's usually over zoom where they're using AI to handle something that they thought was going to just ruin their weekend. And that is something that again, I think a lot of us tend to focus on all the negatives and we don't think as much about the positives and how there's just so much tedious work that it doesn't matter if we can bill for it. We don't want to spend our, you know, precious, one precious life doing those tasks. And what other opportunities could it open up for us if we get through those terrible slog tasks sooner so that we can focus on things like finding new and interesting arguments that we could raise or communicating more with the client or taking additional depositions or who knows? There's a lot of possibilities to still do a lot of really interesting work. It's just maybe different things that we're spending our hours on. So I never get tired of that. I would say another thing that I'll just point out is that there have been a lot of tech waves that have come and gone and had almost no impact on changing the day to day of lawyers' lives and legal professionals' lives. And what's exciting about AI is that it's become such a popular phenomenon that everybody knows about TatchyBT. It's really shaken things up so that we're seeing the largest firms, we're seeing the most conservative institutions start recognizing that, oh wow, we haven't actually made a lot of investment in technology. Whereas our clients have been all using all sorts of technology the last decades and legal really has been stated quite traditional. So for no other reason it really does excite me to see how law firms are really getting shaken up and excited about new ways of doing things. That doesn't happen that often in our profession. Especially because a lot of law firms are managed by partners that might not have grown up with the technology that let's say some of the younger associates have been using. Yeah, I mean, a lot of people still went to law school back in the days, pre-West law and Lexis and all that where you had to get your case cases out of a book in the library, right? And it was all about finding the right page in the book. On the same note, what scares you about AI? Yeah, so I tend to be not super scared of AI. I'm an AI founder, but I would say some of the things that I think should be a point of reflection are thinking about making sure we're thinking carefully about when we use AI, what types of tasks we give to AI. So that's also been a really important part of how I've designed clear brief and our product so that I'll give one one concrete example, which is most people when they think about generative AI, they think about prompting. And that's, in my experience, for working with lawyers, pretty challenging for most lawyers to not only are they using a new technology to try and get it to do something complex, but they're also being asked to do a bunch of complex prompting to get to that result. And that can be pretty difficult. So in designing clear brief, our generative AI tools are not do whatever you want, ask it to do anything. It's a curated selection of specific tasks that we know will work well and give consistent results. So one thing that I don't, I wouldn't say it scares me, but I just would urge more thinking about is I am starting to see some users come to us and say, oh, I want clear brief to tell me the outcome, what the outcome should be of this investigation, or I want clear brief to tell me who's guilty or something. It's really not designed for that. We don't have that as one of our features that, and you can ask chat GPT or other tools to make these sort of very human conclusions. And so I really just urge caution in how we think about what we should be asking AI to handle as part of our workflow. And again, that's all part of choosing products that support the way that you want to implement responsible AI in your practice. Yeah, there's the term predictive analysis. And I think some people over rely on AI to perform some of those tasks. I was going to ask you, are we, so a question that I get asked all the time, is are we on the verge of replacing human arbitrators with AR arbitrators? And then I guess if that starts to happen in any capacity, we have the Constitution, right, where we have a due process clause, I assume the courts are going to be the check on that power. Yeah, it's interesting. Actually, this past week, a new tech startup was announced. I think it's called Arbitus. And their proposal is to have AI arbitrators decide sort of low stakes cases. I do think there are still some unanswered questions about that because I imagine what if someone violates that order, then what, then you have to go to arbitration or go to court. So it may not be as final of a of a judgment as we'd want. But there are, I would say, in other countries in China, they've had AI judges for years now, handling things like traffic tickets. But I still think that for most of the disputes where people care enough that they are going to get a dispute resolution forum to get it resolved, they want the human judge because it is due processes about having that day in court and feeling heard. And so I do question whether a fully completely automated system would give the parties that same feeling of being heard and being respected and listened to and having their arguments fairly evaluated. Yeah, I think it's a really interesting area. Our job, though, really is to show, as legal professionals, to show why human judgment continues to be so important for the law. Yeah, I was at a conference and I was sitting down with a gentleman. You talk about using kind of an AI arbitrator for a smaller kind of consumer level disputes. He had developed a software that was helping airlines process ticket refunds where in the old days, if you had an issue like your flight was canceled or impacted your life in some way, you had to call the airlines 1-800 number. You had to wait a long time on hold and then finally talk to somebody and explain your situation and they'd say, okay, I've noted down your case and I've got to see what we can offer you. We'll get back to you in a week or two. And now he has a software essentially that you just typing you're a propped what happened and it spits out a couple of options. You will give you this amount of money back or this number of miles or this flight credit and then there's always an option if you don't like any of the, I don't want to call them awards, but any of the options that they present you to talk to a human, you can reject all them. But yeah, we're going to probably start to see for small disputes, stuff like that. Now one thing that our CEO, a president, CEO, Bridget McCormick talks about is access to justice and promoting the social justice. And a lot of people are talking about how AI can promote social justice. Could you explain how AI furthers social justice? Yes, actually, I would love to talk about this. We've been doing a lot of work in the space at ClearBrieve. I'm really proud we recently worked with a clinic here in Washington state on MLK Day, where we helped over, I believe it was around 120 to 160. It's on clear exact numbers of pro bono. ClearBrieve recently was used in a really exciting clinic that happened on MLK Day here in Washington state, where we made available our generative AI form filler tool to help hundreds of individuals receive pro bono legal assistance. And it was a huge success. And the idea in line with what we've been talking about is that the AI would do the first draft of this very complex form based on documents and responses that were collected from the client, making it so much faster for the lawyer who was reviewing the form to come in and edit it and correct it and make sure it was accurate. And there's a lot more that we're doing specifically actually with asylum cases and we have a new workflow that's been specifically designed for pro bono asylum work. And we've been hearing amazing things about that from our customers. So that's a big area of focus for us this year. And I believe that it's just a huge opportunity where there are many individuals who want to do pro bono work, but feel like they just don't have the time with their practice. And we all have billable requirements. We all have very busy lives and schedules. AI is very good at saving time on certain types of tasks. And so that's where I'm really focused on and finding ways to organize clinics for helping with tasks where look, if AI can do all of this tedious stuff and just get you that draft, a human in a day can come in and review that. A lawyer can give their legal edits to that and the client can end up with something that's really life changing and meaningful to them and it only took a day. So I think there's a lot of those types of opportunities and that's one of our big focus areas for the year. Yeah, we talk about that at Tripbole how AI can help, especially prostate claimants that don't have access to an attorney or maybe can't find an attorney willing to take their case because it's relatively low stakes can help them with filing their case and drafting documents and whatnot and giving them an overview of the arbitration process and how to understand the rules and all that. Okay, very good. I wanted to add there. I do think it's a win because when legal professionals who also are trying to learn how to use AI, if they can get free training on AI tools and actually themselves get some upskilling in AI and do good and do pro bono work, that's a real incentive to people. They have, they want to do good and give back to their community. They also want to find time to learn AI and it's hard to find time to do everything. So this is a great alignment of interests. All right. Is there anything else you'd like to share today? Just that I'm so thrilled to be partnered with the American Arbitration Association. It's wonderful to see arbitrators who are some of the most respected and members of the legal profession learning new tools diving in. It really is such a sign of progress in our industry to see that and it's going to have a huge impact. All right. Thank you for thank you very much for that and thank you for your time today, Jackie. Thank you so much. Thank you for joining us on resolutions. Today's episode has illuminated the powerful impact of effective dispute resolution and showcase the incredible individuals who dedicate themselves to this important work. Our mission is to explore the diverse world of dispute resolution and we hope we've inspired you to engage with conflict and collaboration in new ways. If you're not yet a member of the ABA section of dispute resolution, we invite you to join our community of professionals making a difference. Your involvement can help us build a more harmonious society together. And if you found value in our conversations, please consider leaving us a review and a five-star rating wherever you listen to podcasts. Your feedback helps us reach more people and continue our journey towards excellence in dispute resolution. Until next time, keep solving problems, building bridges and creating positive changes in your communities.