Video: Commercial Loan Demo | Summary: AI automates data extraction and document creation seamlessly, eliminating manual input and custom integrations.
Video: Running the Interview | Summary: AI-generated HotDocs interview template with prepopulated answers for streamlined demo and testing purposes.
Video: Reading and Populating, with HDAI | Summary: AI-powered tool efficiently extracts, analyzes, and auto-populates answers from uploaded document during an interview.
Video: Template Author | Summary: HotDocs leverages AI to automate complex template creation with seamless integration and robust functionality.
Video: Inside Mitratech HotDocs: AI, MCP and the Road Ahead | Duration: 2170s | Summary: Inside Mitratech HotDocs: AI, MCP and the Road Ahead | Chapters: Welcome and Introduction (0s), HotDocs Overview (75.024s), HotDocs Capabilities (198.66400000000002s), Audience Role Poll (328.04400000000004s), HotDocs Architecture (402.67400000000004s), AI Integration Foundation (508.014s), AI Template Generation (571.484s), Live Template Demo (833.024s), AI-Assisted Form Completion (973.924s), AI Auto-Fill Demo (1198.794s), AI Document Review (1334.944s), AI Security & Privacy (1482.934s), AI-Excel Integration Demo (1529.704s), Document Generation Demo (1691.4691s), AI Integration Recap (1883.934s), Closing Q&A (2035.164s), Closing Remarks (2086.2839999999997s)
Transcript for "Inside Mitratech HotDocs: AI, MCP and the Road Ahead":
Welcome, everyone. Thanks for taking time out of your busy schedule to join us, and thank you again, Jen. So as Jen mentioned, my name is Regatta Belvata. I'm one of the account executives here at Mitratech. Joining me is also Nigel Richards from the HotDocs product team. Hi, everyone. And he will be the man of the hour or actually thirty minutes since that's all we have. And we will be walking through the HotDocs interview experience, offering an AI road map demo here in a bit. Before we get started, we're really just trying to understand who we're working with here today. So we wanna understand your current process. So if you could just take a moment and answer the poll listed below, we We wanna understand what your current process is and how far you're along in terms of, actual automation, AI, and generation. So do you currently draft manually in spreadsheets? Are you using rule based document automation today? Are you using a combination of AI and rule based document automation together? Or do you simply use AI tools like Claude, ChatGPT, Copilot, the like? So I'll give you guys a few seconds. Okay. Hopefully, that was enough time. So now... Okay. So based off the polls here, I see about 27% of folks are using things manually. So pretty green in the space. You do use something like HotDoc, so about 50%. That's that's great. A little bit of both, 22%, and there's nobody currently using just AI. Okay. Awesome. So before we get into the demo, I really just wanted to take a moment to set the stage regarding what Mitratech is and what we're all about. So as we know, in such a highly saturated market, choosing a technology partner is a huge investment in all of your team's time, money, resourcing. And when looking into and cruising the market, you need confidence in not only the functionality of the software and what's being offered to you, ensuring that you're not getting over promised under deliver, but you also wanna make sure that the partner that you will choose will be around and not only five, ten, fifteen years so you don't have to go through the entire process all over again. So we know that this market, like I said earlier, is saturated. It's only growing. Technology is moving very, very, very, very fast. I talk with people every single day who have trouble just identifying, like, where do I even start. So actually, I read an interesting statistic yesterday that showed that, the US labor, department showed that more than half of information sector establishments open in 2020... In 2013, excuse me, were no longer operating five years later. So that means the chances of someone partnering with that sector, five years later, they had to do it all over again. So hot dogs, however, has been trusted for over forty years. Now some of you might be thinking, okay, forty years, that must mean their legacy, they're outdated, they're not modern. I hear that all the time as well, and that is simply not true. So HotDocs is the consecutive market leader in document automation. Today, year over year, according to legal tech hub, as you could see in the top right quadrant, HotDocs continues to be the market leader recognized in tier one, like I said, by legal tech hub. And the reason being is there is not a single document type that HotDocs can automate. So that is due to the robust logic that HotDoc supports and which allows us to support every single type of use case. So I saw from the registry attendee list, there are so many different verticals joining this meeting today, from a large pool of different industries. So we could support everything everything from complex, rule driven documents such as contracts, loan agreements, real estate documents, corporate documents, to also very high volume documents such as NDAs, engagement letters across legal, banking, insurance, real estate, and beyond. So the chances of hot dogs being able to automate a document and scale and grow with you as your needs scale and grow is basically insurmountable. And that's where, however, the next chapter comes in. So with Mitratech acquisition of hot dogs a few years ago, making that investment, we're building on decades of document automation expertise with now modern cloud and AI capabilities. So really the theme here is just to speak to the fact that HotDocs is not only proven reliable, but we're all... We're also modernly capable as well. So before we jump in, like I said, to the AI functionality, I think when I speak to clients every day, there's a there's a huge confusion of should I go with AI or should I go with something like hot docs and automation? Where we see the most benefit is when you infuse the two together. So you need that strong structure and foundation in order... And that proven reliability that already drives results and then adds AI on top of it. So with HotDocs, clients report an average of 90% reduction in document generation time, 50% reduction in errors, giving t... Giving your teams back, time while reducing costly correction. So this increases compliance, increases revenue growth, and ultimately increases time offsetting staffing costs and also increasing compliance. And what makes that possible is the strength of the templates behind those documents. So we handle complex conditional logic so the right language appears in the right circumstances, and your team ultimately controls the approved content down to numbering, spacing, and formatting. So before we move on to basically how hot docs works, another poll here for the for the group. What is your role currently in document creation? Are you creating and maintaining document templates? Are you drafting and assembling documents for clients based off of those premade templates? Do you review or approve documents for accuracy? Do you evaluate, implement, or support document technology? Or do you manage document workflows, team efficiency, and reporting? I'll wait a couple seconds. Nigel, what do you think most of our attendees will say? I think it will be creating and maintaining templates. Actually, only 50% said they used long as more... No. I think a. I think it's gonna be drop b. Let's see. Oh, okay. So 29% are part of the actual creating of templates. 22% focus more on assembly, reviewing, 14%, 25% evaluate and implement or support the technology, and 70% men... Or seven, excuse me, percent manage the document workflows. So this brings me to how HotDocs actually works to keep it very high level. So, again, this is HotDocs as the foundation as the structure. There's two types of users that work within the HotDocs world. The first being the content author. Those are the folks that actually generate templates, maintain templates, and then upload them into the user interface so that the end users can access those templates. Now the content author, everything is done. It's a Microsoft Word plug in. So if you know how to use Word, you're about 80% of the way there. So they go ahead and insert appropriate variables, conditional logic, different robust logic within those templates in order to create that streamlined reusable template for the end users. Once those templates are generated, and this is what Nigel is gonna show us here in a bit, those are then uploaded to Hot Docs Advanced, which is the end user portal. That's where all of your templates will be organized, accessible so that the end user could go ahead and insert and answer the guided interview in order to assemble the documents Yeah. Moving forward. I think it's really important to point out that it is docx. We use Word. And that means that any template that you have, you can... If you know Word, that's 80% of what you need to know to use HotDocs. And we can handle all the styles, headers, footers, margins, all the fonts, outline numbering. It really is one of the strengths of HotDocs. And HotDocs document automation is really quite sophisticated as well with how the the sophistication of the logic that you can use to build the document, but also the questionnaire, the interview that is mentioned, for the end user. Absolutely.... Where I got. Yeah. And I think that's why we built it that way. We we realized that most people use, obviously, Microsoft Word, and how easy it is to generate those templates right through Word itself. Awesome. And that brings us to the main focus of this webinar, folks, is AI. How does that tie into the already built in structure and foundation? So like I said earlier, with so many tools promising AI automation or both, it's hard to know, like, what your team needs, where to start. So what we recognize is starting with a base foundation and structure, a reliable proven structure that works and then powering AI on top creates the most efficient and streamlined system. So think of AI and automation as like a car. So automation, Hot Docs is the engine whereas AI is the added force for it. So Hot Docs, what it does is it proves, provides the foundation, excuse me, your approved templates, rules and conditional logic, repeatable document assembly. So you already have the repeated structure. Now AI is going to accelerate that efficient process. So it's gonna extract source information, help draft language, help you create those templates through implementation, and then also reduce manual data entry. So without further ado, I'm gonna now pass over to Nigel, the man of the... Okay. So I'm... I've got three things that I want to show you today. Thanks very much, Virgo. The first one is how we can leverage AI to accelerate the development of templates. And by the way, this is the one that gets the most wow factor on on live demos if you've seen one. Secondly, we're gonna show you how a... AI can be used to help make filling interviews very efficient. And thirdly, we're gonna be showing you how you can connect HotDocs for document assembly tasks directly to your existing AI tool, whether that's gonna be chat GPT or Anthropic or Copilot. Connect your corporate AI tool of choice to HotDocs so you could leverage AI and deterministic document assembly via HotDocs. So I'm gonna share my screen now just to make sure I get this right. And I'm gonna share the entire screen. And hopefully, you can see that now. Is that sharing? No. No. Is that not share... It is sharing. Yes. We got it now. So on screen, for those of you who don't know about HotDocs, this is the HotDocs author, workspace. And I've only got two templates here. Now before we go into that first demonstration of AI being used to generate templates. Let's just take a very quick look at what it is... What it... What's involved conventionally. So here's a commercial lease agreement, and I've got it open off screen here. So I'm just gonna bring it on screen. Now we don't have enough time today to go into the depths of, of of how HotDocs works and how you you create templates. HotDocs is is... Because you're in Word, you don't have to learn a new a new word processor, And we embed, what we call a field editor, and you can use this field editor to create variables, create, all kinds of conditions. And this is a a simple commercial lease agreement. It might... It's not that simple. You can see there's there's conditional logic. There's placeholders. There's probably repeated sections in places as well. So you can see... You can create repeated sections, punctuated lists, formatting markers, and all sorts all sorts of things. Like I said, if you want to get a deeper view of, what HotDocs does and what you can do with it from an offering point of view, Speak to Regard, and we'll get a demonstration or some... Somebody to help to show you in in detail. This is definitely not enough time, but there is a common saying we use internally that HotDocs is either as deep as the ocean or as shallow as a pool, so it can get pretty robust. Yeah. So absolutely. You could do it sometime so we could show you the full gambit. But, interestingly, this template wasn't drafted manually, although I can. I can go into any of these and see the conditional clauses, and I can edit them and modify them. This template, the one that you see on screen, was created a couple of hours ago by an AI session. Sometimes if I had a bit more time, and I wasn't showing so much, I'd actually do the live the live creation on screen as we wait. Because there's so many of you and I've got so much stuff to show, I'm just gonna show you the chat that Jim and... That created this. So this is simply Claude, and we've kind of taught Claude all about HotDocs templates, what the rules are. And as a result of that, we can create a prompt. So we can give Claude a prompt. So create a HotDocs template called commercial lease agreement to a commercial lease between a landlord and a tenant for a business premises premises. We've got... We've guided Claude about jurisdictions. We've guided Claude about car parking spaces, commencement dates, and all sorts of other stuff that, I'll be honest with you. I I helped Claude help me draft this prompt. I'm not a a lawyer. But you're... You... A lot of you on this call are. You know about the kinds of documents, the kinds of templates that that you need to create. Use existing existing examples of the existing contracts and upload them into a tool like this and ask it to automate it for you. So Claude then goes off and it take... It takes about ten minutes in a normal a normal session, And it generates and it generates all the templates and places them back into my my folder, which is which is what we had open earlier on. So let's just take a quick look at what Claude created. Let's run the interview. So if I run the interview, those of you who don't know HotDocs, you can just test the interview on screen. So just waiting for that window to... It's appeared off screen. So this is the HotDoc interview that was generated by the AI. There's nothing on this user interface that I've intervened in to correct. Now I can do... I can go into author. I can go into the template authoring environment, modify the prompts, perhaps move the move the fields around to make them easier to read. But everything here, including this button, which says, you know, fill with sample answers, because in my prompt, I said, hey. I'm gonna be in a demo later. So could you create a template that you generate for a commercial lease agreement that, will prepopulate with with answers just to make the demo quicker. So there we go. There's some prepopulated fictitious answers. So imagine I was testing this template as an author, a template author, and say, yes. This is all... This all works great. We have I'm sure we've got conditional logic in here somewhere. I'm actually looking at this for the first time. But... Yeah. Yeah. We got conditional logic. You see, is a service charge payable? Yes. All the relevant questions come in. No. The AI, Claude, figured this out based on the the the the skills or the guidance we've given on our template offering plus the prompt which we gave it earlier on. It worked for any any kind of template. And, of course, as you'd expect, when I finish the the interview, HotDocs will take the data that's been entered into the interview, combine it with the business rules in the template, and give me a link to download the Word document. And we'll open it up here just to just take a look at what it looks like. And there... There's a commercial lease agreement. So, literally, this template... And you saw how many fields it has. You saw the conditional logic in the interview. None of it existed more than a few hours ago. If you want to learn more about this and where we're going, this one day not too distant future will be embedded into HotDocs author. So you'll be able to use... Not just go to Claude and generate templates, but go into author and ask author either to create a new template from an existing document, create create a new template from scratch, amend or merge existing templates, and all the time, you can also use the manual tools, the the the the conventional tools to make small tweaks and adjustments. So, Nigel, I wanna ask you a question. So I'm sure lots of folks are questioning that. They're like, okay. Well, if we already have Claude and we could just insert this to Claude and and tell it to do... Prompt it to do whatever you just did, why then... What's the point of HotDocs? Why is HotDocs the important structural foundation to that? So the first thing is that Claude is very good at validation. So you can take your template and it could check that it works properly. AI will make mistakes, by the way. It does. And we all know that. So as author validates and checks, it is also the interface that allows you to test the template before you you give it to your end users. And, also, you could manually add and manually tweak things that the AI hasn't got got correct first time. You can reuse components from other templates and so on. So there might be a day where you don't need the a... Don't need the author tool environment at all. But for now, and I think for the... A good a good part of the future, we've given our customers the choice to use either AI directly and author to validate, test, and adjust the templates before you get... Release them to your users. Yeah. That's a great point. So as someone who is on the client facing side whereas, of course, Nigel's on the product side, I hear all the time that people really are excited to get started with HotDocs and the ROI and the value, that incurs, but they're a little bit nervous Absolutely. Internal resourcing and implementation. So this will just streamline that preliminary implementation process to get your template started and for you guys to get the ROI sooner rather than... Absolutely. Yeah. You bang on. We hear that all the time. Yep. So just to just to remind what we've just seen so we don't keep keep myself on track as much as anything else. What we've just seen is using AI to allow template authors to save time by using AI to create or modify an existing template. Next, I'm going to show you how we can use AI to fill one of those interviews that I just showed you earlier on. So this is where my navigation skills, which, here we are. Alright. So for those of you who haven't seen Hot Docs, Hot Docs event, this is Hot Docs advance. This is the user interface, which not the template offers necessarily, but the end users are are provided with in order to assemble documents. Sorry, Nigel.... See. So sorry to interrupt. We actually had a pretty good question from the panel. A few come in. How does the AI know what data fields are available or should be created because they are missing? Oh, so... Well, you can ask. So it depends on the... You could guide the AI. You can say, I only want the the tenant information to be automated. So just tell Claude what you want to happen, and it will do it. And the AI will just adjust. I cannot... Sometimes, I've found that the AI goes too far and it automates something I don't want automated. I just tell, oh, I see you've automated this section. I don't need that automated. That actually happened earlier on when I had clumsily asked Claude to deal with all jurisdictions, and Claude suggested that might not be very useful. That's to restrict it. So you can have a conversation with the AI to to take things out, put things back in, and adjust things. So you tell AI what to do. The AI, to the best of its ability, will follow your instruction on what should and should not be automated. Awesome. And we also had... I'm sorry. We also had another question come in to clarify, is this AI and MCP to assist the authors developing the forms? It's not that AI helps the users completing the forms. No. So, Van, so we just showed you Yeah. The first part, which is helping developing of the templates, and now we're gonna show you the second, which is completing the form. So now I go back to you. Okay. Yeah. So... Exactly. So let's imagine I need to create a lease agreement. So I'm going to go to, my property for... And so I'm logged in as administrator to this user interface. You've got... I can manage work groups, which is a way of, dividing up the the application into various verticals. So different practice areas can work in different verticals, work in different work groups. You can manage templates and users and so on. Again, if you want to know more about this, speak to Regard, and she'll get a demonstration together for you. Today, though, I'm gonna show you how we can use AI to help a user fill the interview using a document that they have from somewhere else. So we're gonna create what we call a work item, and I'm gonna call this a b c corp. Just just... And I'm going to type the name on top, select the shopping mall retail lease agreement and launch the interview. So we're not no no longer an author. Let's imagine that we're somebody in front of business who's tasked with actually creating a retail lease agreement for a customer. And normally, I would go type through this. I'd type the landlord's legal name, type through this, go through the whole interview, which is great, and it's far quicker than drafting a document manually. I could do that. But we've added a new feature. By the way, this isn't mock up. This is real. I can click the AI auto fill feature. And I just realized I didn't have the document ready, so just hold on a second. I have a document here, which is a term sheet. So often in in... When you're creating a contract, one of the earlier documents that get created, sometimes people use hot docs to do that. But sometimes you get a summary of all the major terms in what's called a term sheet. This is just one use case. It could be, a transcript from a a meeting. It could be copy... An email. It could be anything. Basically, we're taking unstructured data from some other source in the form of a document, dropping it onto the interview, and we're going to ask Hot Docs AI, Ares AI autofill, to read the document, read the questions in the background interview, and then populate all the answers. Now this is quite a decent sized document. I should have shown it on screen earlier on. And as you can see, there's lots of questions in this interview, so it does take a few moments. But within... Oh, there we go. So So we can see... Remember how many questions we had answered earlier on? We had... Every one of these dialogues had lots of questions on, and they've... The vast majority of them have been answered by the AI tool using that document I uploaded. We can see which doc... Which, where the AI has filled in, so we make note to review it. Obviously, AI can make mistakes, so it is important to review the data. We've got some fields which haven't been haven't been answered, so we'll fill fill some of these extra extra fields in. They didn't pick out the... Perhaps The states, the shopping center state wasn't actually in the terms, and we'll just make sure it's enclosed regional mall. And, yeah, I happen to know there's 56 in that particular area. Right. There's there's x, y, zed corp. So there's a... It doesn't capture necessarily all the fields, so it could do if you had the right data. Change of use, we'll just say no, and then it is more dynamic. Remember, this template was generated by AI as well. We'll answer no to these questions, which presumably weren't in the term sheet. And now we can see we've answered all the questions in the interview in a fraction of the time it would have taken without the AI. We can click finish, and then the... This is where the the AI is collaborating with the rules based document assembly to ensure that we've got precise sanction clauses in the document. We can open up the document, we can download it, and there we have the the the shopping... The retail lease agreement opening up in Word. Nigel, let's see if we can check. Very quickly here. Question from the panel. Will the AI autofill document uploader support markdown files in the future? So the answer to that question is yes. At the moment, it's, it... The... At the moment, we are... We're we're just doing PDFs and DOCX, but we will be supporting images. So So imagine taking a picture of a passport or imagine you're doing trusts and wills and you've got a previous trust or previous will. You can scan it in or take a photograph, and it'll prepopulate most of the details before you then create a new will. So we'll be supporting other formats including just cutting and pasting text. So the answer to that question is yes. Okay. Another question, is the AI a closed interface? I... E. Are there internal protections in case docs have client identifying data and that info is not shared to train the overall LLM? So it's absolutely closed. There's no there's no training. There's no... It's not shared. It's it's completely, if you like, sandboxed. So there's not... Should be no worries there. Fantastic. And as we come on the next five minutes, I think the next question from the panel, perfectly transitions us over to MCP. Someone asked, can we link this feature to our proprietary I... AI system so we could train them to improve accuracy? So I think that is... Absolutely. Yeah. So just as a reminder to everybody what we've seen, a reminder to myself as much as anybody else, otherwise, I'll go off track. We've seen AI creating a template. We've seen uploading a term sheet to create a, a commercial lease agreement. But what... The next thing I want to show you is connecting AI to your own tool. Your own AI tool, whether that is author... Sorry. Whether that's Claude, whether it's ChatGPT or Microsoft Copilot. Now the way I'm gonna show this is by using, an AI connected to an Excel spreadsheet. So lots of the vendors, including including anthropic, have created desktop tools and desktop LLM clients that connect directly into Word. This is not the only way you can use it. You can then just open up Claude or open up Copilot and talk to Copilot and say, could you create a a new a new interview prepopulating? So this spreadsheet consists of a number of rows, and each row corresponds to some data, which is gonna be used to create, like, in each case, a commercial lease... A commercial loan agreement in this case. So I've asked in the prompt down here, just to save some time up, just create a prepopulated interview for each row in this spreadsheet that represent commercial loan agreements. I'm just gonna click that button. And, what what should be happening now, it's gonna figure out what's what's... Yeah. It's finding the correct connector tool, so hopefully it'll choose the right one. And it's... Because I do have quite a low... Quite a few connectors. It's figuring out that I want to create documents. Let me just give that a few moments. While we're waiting for that, Julie... Sorry. I was just gonna address a panel question while you were, going through that. But... So Julie asked, has HotDocs moved to AI only? No. So HotDocs could be used as it's been foundation structure, but we can also turn on the AI functionality. So we give all of our customers the choice. Go ahead. Sorry, Naida. Yeah. There's another one from Julie in Team Connect. Would you select generate documents to learn HotDocs? So, that's... Doesn't... I don't know if, Julie, if you mean that to be an AI question, but there is already a connector in Team Connect for generating HotDocs documents, but... And there'll be more AI stuff going on there, but that's for a different webinar. So I'm gonna ask... I'm I'm being asked by the AI, in this case, Claude in office, to confirm that I wanted to create a work item. So I'll go ahead with that. So it's just taking a few more seconds. There's quite a lot of data in this spreadsheet. As you can see, there's there's quite a few rows. And also live demos take a little bit longer than I'd expect. While that's loading, another question. When users are uploading and using AI to complete the form, if they are using a mobile phone, is there anything that's different interface wise with these features? Yeah. So... Oh, interestingly, that interview you saw is mobile friendly. So the answer to your question is we're not quite there yet with the fully responsive... Of... Responsive UI, but you will be able to use AI to upload an interview. And I have no reason to believe, that's such a good question because I haven't tested, that you can't use the tool that I'm showing now to create work items from your mobile phone. But that new interview, that interview is mobile friendly, and we're doing work to make the whole of the hot docs interface... I mean, it still... It works already, but to make it properly responsive on mobile device, the answer is yes. So we've been told by the AI that it's created new work items. So let's take a look at Northwind Timber, and we can open that link direct from the AI, and it's taking me to HotDocs. There's a Northwind timber, work item created. We we asked for it to prepopulate the interview, so let's click launch from the data in the the spreadsheet. And you can see in this particular spreadsheet... I mean, I did design it, so it had all the data. The AI has picked out all the data, so there's nothing I can just review review the the... What what what the AI has picked out from the spreadsheet. I didn't have to create any, custom integrations here, which is great. It would work from any document, a Word document. I don't need to use a document. We'll get to that in a moment. And I click finish, and I've created the loan agreement for Northwind Timber, and I can download that document. So no typing of data into an interview, just reviewed it. No complicated connectors or anything like that that have to be bespoke for that particular spreadsheet. No mapping of variables because the AI does it for us. And somewhere in my screen, there must be a Word document that's opened. I think I clicked download. Let's have a look. Where did I put it? It's probably one of these. Sorry. I should've... I didn't I didn't download it, did I? Let's just click that and then open it as a commercial loan... There we go. That's the... There's a commercial loan agreement being created. I'm not gonna vouch. I'm not a I'm not a proper banker. I do know there's some banks here. By the way, this works in other languages. So, the other day, I was on a demo, and I used AI to create a Spanish language. We have customers in Germany creating German language templates as well, but there's our there's our final document. So just to remind us before I hand back, actually, just to remind us what we've seen, and I'm gonna hand back to Regard now, Regard... Oh, I gotta stop sharing. So it's my fault. So I'm gonna stop sharing. We'll go back to this. We've seen you... AI to create templates. That's item number one. And it isn't AI only. You can use a combination of traditional tools of creating variables and and, you know, create hand typing clauses. So it's a combination of both if you want. You can save time. You can upload documents in the interview, like a term sheet to create commercial lease agreement or any other kind of document, and you can connect your desktop AIs to any any any other, to hot docs directly to prepopulate interviews. And if we go to the next slide, because this is an important slide I wanted to show. Do I click next? Oh, I've gone faster live demo. It's... Oh, previous. That's me navigating wrong way. Now I'm messing this up slightly. Oh, this is a slide I want to get to. So what I showed you in this... What this diagram is meant to show is is users, these these heads down here, interacting with an AI. In my case, I I use an Excel spreadsheet, so there's documents over here that feed into the AI. And I'm actually in the document when I'm using the AI using HotDocs, the top doc deterministic assembly to generate all these documents. And we know all the clauses have been sanctioned and all you're using AI for is to help you fill the interview. But your AI could be connected to messaging things like Teams. It could be connected to emails. It could be connected to other things. So your organization... I even have a tool that helps me read emails in the mornings. I get so many. I don't know how to help me filter them. It might come up in email. I could use that email to say, oh, there's an email that's asking for new loan agreement. Could you then just create the, use HotDocs to generate the the agreement for me? So I don't want to use AI to actually draft it for me generatively. I want to use a HotDocs template so that I know every clause is accurate and compliant to our organization's criteria. And that really, the the use of AI, I get all... Every demo go on, I talk to different customers. It... It's really how these customers combine our services with their own AI tooling to achieve the road... Results they want. That's where the real power is. Anyway, I'll stop talking. I always talk too much. I'm gonna hand back to, to Regard now. I think we're almost done. I'm just gonna just say thank you. Yep. Thanks, Nigel. Yeah. I wanna be really mindful of time here, so let's wrap up in the next minute or so. Just gonna address the last two questions here in the panel. Any available quick response, any availability limitations for software versions and... Or on prem hosted. So we do offer on prem hosted solutions. Yeah. How that's good on prem. Happy to discuss that further with you on a call. What MCP server tools will Hot Docs developers have access to in the near and future road map? So the near future is the one I've showed you, where you could create work items to prepopulate interviews, but we will be expanding that as we go forwards. Fantastic. And then, Katrina, I will get you involved, for the TeamConnect related questions. I'll relay those to your CSM, and get you on a formal discussion so we can really talk through some of those outstanding questions, regarding your Team Connect environment. But, yeah, I just really wanted to say thank you guys so much for joining and attending today. I know we ran a little bit over time, but hopefully, the whole goal here is to understand that HotDocs will remain the document engine driver, and AI could be used as horsepower if you so desire, but you can always, leverage HotDoc standalone without AI. Yeah. Talk to people at home with AI all the time. Hey. Just twenty seconds. Somebody asked, can you use this AI tool to update existing templates and modify them? Absolutely. That was a question I just missed there. I didn't have time to demonstrate that today. Yeah. And and please feel free to email me. There's... I think the email here is on the last page and then the QR code. If you would like to set up some more time to discuss Hot Docs and how it could fit potentially in your environment or just learn more, happy to happy to have a discussion with you. Alright. Well, it's been... Thanks for watching. Bye, guys. Bye.