Unith Ltd ($UNT)

Yes. Good day, everyone. A very warm welcome to this webinar, which will give you a bit of a business update, product enhancements, but also a strategic outlook. My name is Sytze Voulon. I'm the Chairman of Unith for those of you who don't know me. My colleagues will do a quick introduction themselves, and then we'll start with Scott giving us a quick business update, and we'll flow through the presentation. Very important at the end, there will be questions and answers. So please do make use of that opportunity. Otherwise, it's going to be a one-stream information towards you guys, but it is actually more interactive and interactive is what we are all about. So it will be nice actually to see lots of questions that we can hopefully answer to your satisfaction. So with that, over to you, Scott, to make a quick introduction.

Thanks. It's -- my name is Scott Mison. I'm the Executive Director of Unith. For those that don't know me, I've been on the Board for about 5 years, and I'll be delivering on the financial aspect today as well as a business update as well. Do you want to put that on, Rakan?

Yes. Just checking. Hi, everybody. This is Rakan. I lead the Digital Human division at webinar -- Digital Human division at Unith. Just before we get started, Gaby, I wasn't sure if you also needed to make a brief little introduction from your side.

Sure. Yes. So I think Sytze has done that already, Rakan, but thank you, everyone, today for joining us. Just a quick reminder before we go into the formal part of the presentation that the session this afternoon is being recorded. And as Sytze said, there will be opportunity for Q&A at the end of the session. [Operator Instructions] Thanks, Rakan.

All right. No problem, Gaby. Scott, I'll pass over to you just to get this started.

Thanks, Rakan. So thanks, everyone, for joining today. And we'll just give -- as Sytze and I mentioned, we'll give an update for the last quarter and also give a business update on the product and also the strategic view of what we've got coming or looking to do into the future. For those -- the shareholders or those investors or new investors that are joining today, Unith is an ASX technology company listed obviously on the ASX, which specializes in AI humans and conversational design. We've built interactive agents for customer engagement, education and entertainment. And we've got 2 divisions, one being the subscription division, which is we use the digital human technology to -- for their AI content and they've got their own web applications and also the digital human, the B2B division, which is we design digital humans, deploy and develop them. Next one. Thanks, Rakan. So as I mentioned before, we'll go through the Q3 FY 2026 financials. We launched those on the 30th of April. Obviously, it's an Appendix 4C, which shows our cash flows. Those 4Cs, I guess, look at only the cash receipts that you received during the quarter. During that quarter, the quarter 3, we received just over $1 million in cash. However, the revenue side was about $1.26 million. And the difference has really been the timing difference between collecting some of those accounts. from our clients. There was 2 particular ones that we collected in the first week of April, therefore, hence, the difference between cash receipts and the revenue. You'll also note that we -- at the end of April as well, 30th of April, we have accounts receivable of roughly about $800,000, which will be coming in over the next quarter and 2 to 3, 4 months. Also on our balance sheet, we've still got our investments within AudioStack and Unique, which are carrying a fair value of $3.9 million. With those 2 investments there, obviously, they're unlisted. We have had the opportunity in past quarters to divest some of those investments, particularly in AudioStack and in the next -- probably the next quarter being July to September, we'll probably look to divest some of those investments as well to bring in some cash into the company. During the quarter as well and just before the quarterly was announced was -- just post the quarter, sorry, and just before the quarterly was announced, we signed off on a finance facility, which was for an extra $1 million. However -- so what that has done is we refinanced the original $1 million that we got in December of last year, refinanced it into a $2 million facility. And now that $2 million is not due and payable until the end of October. So originally, I think our first $1 million was to be done by the end of June. We've -- as I mentioned, we've refinanced that to extend that out to the end of October. So no repayments are due until the end of October. Even with the $1 million, that reflows to that date as well. The facility allows us to give the continual development to the AI platform. The industry is moving so quickly. So we still need to enhance our products as we go along, both on the digital human and the subscription side. We're also looking at -- which we've announced as well, we're looking at some great opportunities, strategic acquisitions that really complement and Sytze will go into it further into where he sees with the vision of where we see Unith going into the future. But there's some really exciting opportunities out there that we've notified and really doing some -- we'll do some DD on those opportunities to see if we can deliver and bolster the company to really give it a good chance to move things forward. Part of the quarter, it's not really a financial thing, but the ISO accreditation we completed. That was a long process, a big process, which took a lot of time and effort from our -- especially of our tech guys based in Barcelona, which did a fantastic job. We went through 3 audits, 1 internal audit, then you have 2 phases of external audits, and then they give a recommendation whether you comply or not comply. We've been very fortunate that we've been given the letter that we have been recommended to get that accreditation. And the auditors are just from A-LIGN Compliance & Security who did the audits. They are just finalizing their reports and looking to give those certificates hopefully in the next couple of weeks. So that's a great -- I guess, it's a great way for us to ensure that our prospective enterprise clients or those are clients who want to make sure that we're compliant and really have that demand for high compliance, they know that we are ISO accredited. So their data is secure, and we can attempt to get those high revenue growth clients. And I'll hand that over to Rakan on the financial side, and he'll give an update on the digital humans.

Thank you, Scott. Yes, so I always think it makes sense to start by just revisiting where we are today, both from a platform perspective as well as a usage perspective. And the best place to start is, and I know Scott touched on this a little bit, but just reminding everybody what exactly is the digital human and why they're valuable and why we believe they will become ubiquitous across businesses and organizations in the future as they look to engage with their customers and their employees. So digital humans are engaging scalable AI avatars. They connect with people, they assist with people. They, in many cases, even convert people if those people are potential customers. And they do this anytime, anywhere and in any language because they're completely digital. One of the key benefits, at the end of the day, we are tuned to engage with human-like faces, human-like people with humans. And just by the fact that the digital human comes across us humanly, it improves the interactions that people are willing to have with AI. I mentioned this they're always available. They speak any language. You can easily customize them, make them look and feel like your brand. Super critically for many businesses is the data that they capture when they speak to people and what businesses can do with that data to optimize their own processes to even retrain or reeducate the digital humans. They're integrated -- our digital humans, as I'll show you soon, are integrated across a whole range of platforms and being used by a whole range of different people. The digital human in many ways is just the face, the power behind it is the brain, and we have all sorts of -- they have all sorts of configurability on what that brain can look like. They could be super, super specialized to solve one very specific task or they could be very broad. They could be better at languages or better at science or whatnot. Yes. And again, at least the way we think about digital humans is they have to be scalable in a way that does not consume a huge amount of resources so that businesses can actually adopt them in a way that's commercially viable for them. And I think Scott also touched on this a little bit. And I'll speak a little bit more to this in a few slides' time, but we see so much desire, and I use desire because it's a willingness still to use digital humans in health care, in learning, in human resources, in onboarding across the board. And part of our challenge is to figure out where is the best place to really start with, yes, where we really build the value proposition of a digital human. I mentioned this slightly earlier that we really do believe digital humans will be absolutely everywhere. And when I say we believe this, we believe this as a team and Unith in the market reflects this view. If you read and there are some links in this deck, which people can look at. If you look at how the market look at digital humans today, it's anywhere between $7 billion and $10 billion is the market size of digital humans. And this is largely the type of digital humans we are talking about. The view is this is going to grow anywhere between [ 1,500% and 3,750% ] over the next 5 to 10 years. So not a very huge time. And from our perspective, we are like -- we are at the beginning. We, in many ways, feel that we will do 2 things. We are going to grow at a rate similar or more than this growth projection is here, both from a revenue perspective and from a market capitalization perspective. And that, in many ways, at least from my perspective, is a conservative view because we don't want to grow with the market. We want to eat into the market and then tap into the fact that the market becomes -- the market becomes more ready and more willing for digital humans. This little strategic takeaway at the bottom right, I always like to remind ourselves this that value is increasingly captured not by having a model, but deploying it effectively at the point of interaction, and that is very much what the digital human does. It is putting the digital human at the interface between the AI, the LLMs and the people who actually interact with these types of AI. I feel something to be very proud of is a fact that we have been live for not a very long time. I think -- if I think through -- if you look at this chart at the bottom right, maybe mid-2024, we really started to make our platform broadly available to users and to customers. And the reality is we've proven that the platform is established, it's stable and it's scalable. To the tune that we've successfully like executed over 2.7 million sessions. So these are moments where people are willing and can speak to digital humans. It's doing what we say it will do and the fact that if you look at the data and you see all the different languages that people are speaking with digital humans, it's going across the board. There are over 30,000 digital humans that have been created. These are huge numbers. So these are actually digital humans that somebody has thought about in the sense, I want a digital human to do this and engage with people in this way and use our platform to create them. And they're running all the time. I had a chart I wanted to show, which was like from Monday to Sunday, basically, it's spread across the day -- it's a spread across that -- our platform is being used across every single day. But oddly, Tuesdays is like significantly more than any other day of the week where people tend to use digital humans. I have no idea why. But again, it's just interesting to be able to see that stuff. On the bottom left, again, just to reemphasize the global languages piece because digital humans are deployed literally all over the world. And if you see this dark chart here, this is actual heat map of users who have spoken to a Unith digital human in the -- over the past -- like since we've launched this stuff. And it goes literally, there's almost no country here that people are not actually speaking with digital humans. The chart on the right is the -- so this is like the heat map of where these 2.7 million interactions are happening. And then the heat map on the right is the people who are creating the digital humans, where are they coming from? So these are like the administers or the employers of digital humans. You could think of this as like our customer base. So there's less coverage of countries. But again, still the -- one of the key takeaways is we have a global platform where our customer base is global, and we're every day dealing with people from all parts of the world. Yes. So again -- and with all of this usage, one of the difficult things with so much data is making the right decisions, having looked at this data and seeing what it tells us. And again, taking a step back, it's clear to us like when we -- when I think like what are we very proud about is that the platform really is scalable. And we've seen huge peaks at certain points in times of usage, and we've seen the platform be able -- be able to support this. And we know that if you add even the words like stability or scalable and stability is the reason we've won businesses in the past because other people who are trying this don't really have a platform as mature and ready as ours. This is probably the piece on the good side here that I feel like is the greenest of all of these boxes here. It's very, very clear to us that a great user experience, UX is key to adoption. And this goes beyond just having an intuitive user interface so that people like understand that you've got to tap on a microphone to activate your microphone, et cetera. So they understand how to speak to the digital human. This is a piece that we know needs to be improved, needs to be thought through the user experience. But in there, there's also the fact that the digital human needs to be fast enough. It needs to feel real. Latency is double-edged sword. You make it too fast, people complain. It's too slow, people complain. So finding the right latent -- the right delay that is acceptable and feels natural is something that we've very quickly realized we need to work on, and I'll show a little bit of how we've addressed that. The visual appearance, it needs to look beautiful. And I say this in all senses of the word, in the sense, that not only -- is it not only the face, but the environment that the digital human is in really reflects what the business who's employing the digital human is trying to do. And I'll come on to this a little bit more, but like emotional control and expressiveness directly is -- I'll speak a little bit more on this, but it's something that a lot of businesses that we are in active conversations for -- in active conversations with today are looking for, and I'll explain a little bit why later. We've also -- and I don't know whether it is unfortunate or fortunately that we've realized that the people who get the most value from our product are people where we've sat with them. We've understood their business problems. We've understood their pain points, and we've helped set them up. So we've almost worked with them on a project level basis. We've done the initial deployment for them, and then they've adopted the product. And we very quickly see that their usage is much higher. Their desire to do more is much higher. Their desire to do more project work is much higher. So we are figuring out a way that we can structure ourselves to support businesses who need a helping hand in getting started with digital humans. Scott touched on this. I'll speak on this a little bit more, but like security and compliance, there's a certain type of business that if we don't tick certain boxes will not work with us. And this is a piece that we've invested a lot and we've thought a little bit -- a lot about it. We feel our platform is very, very secure, but we've not had perhaps all of the documentation and proof points of how secure this is. And going through the ISO process, Scott mentioned, has helped us a lot on this, just doing a quick time check. And then it is -- this is a difficult one for us, but it's something that we need to be wary about, like how you price digital humans is still a bit of a question mark because it is early market segments, early adopters, and they are conscious of pricing because they are not necessarily fully sure of what the return on their investment would be. From my perspective, this means being able to manage our costs operationally well, but also having a flexible enough pricing model that helps people come on board and grow. So again, these are just some of our takeaways. Accounting for all this usage, we've been hard at work, I believe, executing very well on delivering, I'm going to call them like point product enhancements that directly address lots of these features -- lots of these feedbacks that we've had. First off, and I believe we've announced this, and I'll play a short video, but -- our digital humans are remarkably fast in how they respond to questions, whether the answer is a little answer like yes, no, maybe or it's a long, lengthy answer where it's gone to fetch the information from some third-party system. We've done this by building -- designing and building our own proprietary streaming technology, which allows us to have control over the costs in doing this stuff. So we don't need -- we're not dependent on other streaming technology out there that may -- that basically eat into a lot of our margin if we were -- as we sell this. So again, the performance is great. The way -- I've not touched on this, but the way it lets us price actually differentiates from others in the market. in the sense that we don't worry so much about the length of a conversation. Actually, we encourage lengthier conversations, whereby everybody is billing by duration of conversations or everybody tries to bill by duration of conversations. And then most interestingly is just the fact that it all of a sudden allows many new types of use cases. And I'll play a short video here. Around 8 type of use case that we are seeing -- we are seeing great adoption of this specific use case. I'll just play this for a second. [Presentation]

Digital human coach could enhance user experience significantly. How would you explain to a nontechnical client what a large language model LLM can and cannot do?

I say that an LLM is like a very advanced language assistant. It can understand and generate text, so it can answer questions, summarize info or give suggestions. But it doesn't have true understanding or emotions, and it relies completely on the data it was trained on.

That's a clear and approachable explanation. It balances technical and user-friendly language. If a prompt isn't working well, what would you try before concluding that the model is bad?

We've got 2 technology partners/customers of ours that are building full-on HR interviewing solutions on top of this product, and they're doing very well. We're seeing the number of digital humans being used to interview people growing. We're very happy to see them -- to be seeing them progress on their journey on top of our platform. This is another kind of more, I would say, tongue-in-cheek demonstration of me just showing what else having a low latency experience brings. It allows us to be much more creative in how we use the microphone features. You will see that in this demonstration, I'm actually speaking with digital -- with Scott. And I'm being a bit rude to Scott. I'm kind of interrupting him, asking him to say things, stopping him, et cetera. And again, the point in this is that with streaming technology, you can almost barge into the conversation and stop the digital human speaking when you're not happy with the digital human. As always, good to see you, Scott. How are you doing? Curious about that.

Thanks for asking. I'm doing well, and I appreciate your interest. We're excited about the progress at Unith, especially with our recent launch of streaming avatars. How about you? Is there something specific you'd like to know about our latest results or products?

Why streaming avatars? Why is latency important?

Streaming avatars are crucial because they significantly reduce latency making...

I already know this. You know I know this. So maybe tell me about something to do with just keeping systems secure at Unith.

At Unith, we prioritize security by focusing on compliance with standards like ISO/IEC...

Can you tell me this in Mandarin, please?

[Foreign Language]

Scott, I was just joking. I don't understand Mandarin, but I'm glad to see that you can speak it if we need to speak Mandarin. Tell me something else, tell me what excites you most...

No worries at all. I'm glad to showcase my capabilities. What excites me most right now is the rapid growth potential we see with our streaming avatars and the advancements in our technologies and innovative...

Scott, how about ISO certification?

Our ISO certification journey is an exciting step for Unith. We successfully completed the external assessment for [ ISO/IEC 37001:2022 ] and received a recommendation for certification. This achievement...

Scott, that's all right. Thanks for that. Nice speaking to you. I want to get back to this presentation. Sorry about that. I think I had a bit too much fun. Maybe actually listening to this demo today could have done lots of -- could have done lots of the update for us in that form. Okay. Another thing that we've not really announced, and we will -- we are thinking through how we announce this more broadly at least to the market is the fact that we are really focusing on putting digital humans where our users and where our customers are and making it easier for developers to build really amazing products on top of digital humans. So we've launched an SDK, which is a software developer kit that ultimately lets developers decouple our user interface completely and build very fascinating types of user experiences, put them into different types of business workflows so that they do cool things for businesses, value-adding things for people -- for businesses. We've launched 4 of these. There's a React Native one. We have a React -- React Native is for like native iOS and Android apps. We have a React one. We have a TypeScript one. And we have more of just a regular JavaScript widget that lets people mainly think about the web and web experiences customize their digital humans. And why is this cool? Again, this -- maybe I won't show the full thing on this one, just to not bore everybody. But as we lodge this presentation, you'll be able to watch these -- you will be able to watch these videos in their entirety. So here, we have -- if you imagine, this is a digital human again, that is kind of coupled into a business opportunity or lead capturing form. So it kind of sits side by side with other technology, which is a form that captures information about somebody and then uses that to inform the digital human of what to speak with. The very first thing that might not come across when I hit plays, you'll see that I type in name Jimmy Bean and then I hit go and the digital human then knows that it's speaking to Jimmy.

Nice to meet you, Jimmy. Welcome to Unith. I'll help you through a few quick steps, so we can understand your project and connect you with the right team. To start, could you describe a project we could help you with?

This is really cool. I'm actually looking to deploy something just like you to support my customers when they're looking for new information about my business.

That sounds really interesting. Using a digital human for customer support and information makes a lot of sense. Could you tell me a bit more about the kind of information your customers would typically be seeking from this digital assistant?

They would be trying to understand price. They would be trying to understand different features and functionality that my company offers and they would really just be looking for consultancy and guidance.

Got it. That's a strong use case, helping with pricing, features, and consultancy can enhance customer engagement. How developed is this project on your end? Could you tell me your company name?

We are called Bean Factory.

Thank you very much for the information. We'll be in touch soon.

I do want to speak through this point. So what's interesting in these kinds of solutions is the digital human asks a bunch of questions, understands the customer, the user, whoever it may be, does some form of qualification of this person, summarizes the conversation it's had, sends it through to an internal system says that, hey, I've spoken with Jimmy Bean in this case from the Bean factory, looking to help to be able to answer questions to their customers about pricing features, et cetera, understands all of this information, gives it a score and then decides what do we do with this person. And because they've seen that I'm really not a real person, I also said that I didn't have budget in other parts of the conversation, which is cut out from this. The digital human then sends the user an e-mail saying, hi, thanks for speaking with us. We've understand that you're trying to do this. We actually think, in your case, the best thing to do is to go through the free trial on the platform. So rather than taking up our sales team's capacity to have a conversation with this person, it's just sending them to the free trial, the user will go in there, see whether they can see what they can do with the product, see if they can build a little business around the product or application with the product and then make their own decision from there. We've also -- just as a side point, we've also changed the trial terms for our digital humans. So if you go into the product today, you have a 2-week free trial after which you have to make a decision. So again, that becomes a point back in that if they actually use it and they did actually feel they could, I don't know, in whatever case, all of a sudden do have budget, which they didn't want to tell the digital human, it would then connect them back with us so that we could have this conversation. I'm speaking this in the sense that this is what we are doing with it as a showcase, but this is actually use cases that the digital humans are being applied -- that our customers are applying digital humans themselves and building these kinds of use cases for themselves. A lot of these features, I have to stress are coming from like understanding what product users are telling us indirectly through their usage of the platform and some -- and let's say, some forms or feedback forms that we've sent. But the reality is we spend a hell of a lot of time working with our existing customer base and prospective clients who are looking for things. And in doing that, there's some big learnings for us. And this is defining a lot of our longer-term planning, both from a product standpoint and a business and a go-to-market standpoint. So very quickly, I think it's very, very clear to us that where we see the immediate -- and I would just call this like willingness to spend to solve problems are really in any businesses who are struggling with human resourcing constraints. And a big part of this is actually the ability to train staff using simulated conversations. So businesses with large sales teams, businesses with many field workers, hospitals, pharmaceutical companies that spend a lot of time and effort in actually training people. It's like one category or where they hire people to do repetitive tasks. So -- and again, it's very, very clear, and you can work with them. It's very -- it's much clear to build a business plan to adopt digital humans to address this because the return on investment is very, very clear right from the start. So this is where we will be focusing on and what we are focusing on now from a lot of our new marketing initiatives. as well as a type of business that we are calling it like white labeling AI, white labeling tech. So these are technology companies who already have a customer base who are looking for something new and value-adding to take to their customer base, and they are self-sufficient. So they don't have a dependency on us. So it's like a pure licensing model. And we see a lot of success on this today, like actual customers that are doing this today and ones that want to work with us, and we're working through. And the part of the challenge just openly speaking on this is part of the challenge with this one is the revenue takes time to come through in the sense that they need to succeed themselves and it takes -- it just takes time. Like they need to start getting their customers to adopt the digital humans. They need to start seeing value in it. And they slowly, slowly grow, and we hope that -- we trust that growth will compound, whereby the top category of human resource constraints, there's budget and revenue to be spent right at the start. And this comes back in with us helping them do the initial project to get them started. And that's kind of how we get immediate revenue from businesses that we work with. I'll -- again, this is just things to bear in mind, and I'm sharing this and we are sharing this just so that -- like our investors and the market understand some of the challenges that we learn and what we're doing to accommodate this is it's still an immature market in many ways where people are figuring out the value proposition. Which is why we are very convinced of this being -- this segment very much in the training and enablement of people space being where people are willing to spend right now. we have to realize that we are working in a market where customer expectation is super high because they see all these beautiful tools and technologies out there that can be distracting and it's something that we need to learn how to live with and show the value of our platform, which we do once we get to a certain level in the opportunity or once we're working well with the customer after they become customers. Role playing is standout use cases, extremely powerful use cases, especially for larger orgs who do -- actually do a lot of role playing themselves, but want to be able to scale that much more. In the role playing category, emotion and expressiveness is clearly missing from our existing solution. It's the #1 request that we have. And then Scott touched on this, but compliance the problem with compliance is not that it takes time. I don't mind having 10 opportunities that are taking time. The time we have to spend on our side answering all these questions. And that's why we are very excited about what Scott mentioned. I'm just thinking of a most recent due diligence that we just went through, we literally sent back their form is like, are you ISO compliant and we just ticked it and we sent back this letter that you see on the right here saying that we are there. And it just streamlines everything. And we have all the policies already written now and approved that we can send because people ask us, show you have this policy, show you have this policy, show you have policies you cannot imagine exist, exist, and we didn't have any of this well documented. So we've gone through this process. We're super happy that it's behind us, and we're super -- and we're looking to capitalize on the fact that not many companies in our place have actually gone through and been able to get this in place. So that's one thing. Second thing, more product-centric. I'm running a bit behind, but this is the most exciting part of the whole presentation. Let's not -- to me, it's super, super simple. We need to do everything we can do to be a category leader, like we need to be at the top of the whole pile. We don't want the AI noise to slow down opportunities for us. We have a small but mighty research team. We would love it to be bigger, but this is what we have. But these folks they are pushing the boundaries very, very quickly, and we're very happy with the output that's coming out from this team. Our digital humans, the realism needs to be better. We need to have expressive digital humans. It needs to all be done in a way that is performance from a latency perspective and also from an operating cost perspective. And we are also a little bit thinking a little bit more down the line around completeness going beyond just the face and doing more. And I'll show you some of the output that's coming out from this team. One step back actually, we decided late last year that we needed to kind of revisit the whole lay of the land. And we went through a 3-month initiative where we basically looked at all the state-of-the-art models that are out there. And what's actually interesting is the research that is going into this space goes right back to what I said at the start of the market growth for the digital humans space. The number of research papers that are published year-on-year is growing at a rate that just shows the amount of investment that's going in and will continue to go into this. So we've looked at this and the output of this has been that we've developed a novel unit proprietary real-time audio-to-video model. We have a demo that's built on our Unith existing infrastructure, and I'll show you some pieces of it now, just the video output of it. As a next step, we are planning to bring this into production. We will have all like go-to-market plan on how we bring it into production, how we announce it. It's got its name. I'm not going to mention the name over here. Otherwise, the team will kill me, but it's going to be version and the whole intention of the versioning is just to set the tone for where we're going in the sense that you're going to see a V1 come out. And we have very big plans for what that could be. We will also soon be publishing more research papers. In the end, we are Unith research labs. We do invest in research, and we are going to start building our credibility in the research community as well. So there will be a research paper comes out soon that is a bit of a survey on what the lay of the land looks like. The model we have yet will be our most expressive model yet. It's coming soon. You will see very quickly how much more expressive it is than the previous versions. Not only the -- you will see that the whole head is expressive, the whole head moves. We've done it. It's already plugged into a conversational system, so you can actually speak to it. So it's kind of our big in the -- in our 3 dimensions of how it looks, how fast it is and is it conversational in those 3, it's kind of right there, it's addressing all 3 of those. And then it's really powerful in that it can be both generated from a single image, which will be less quality or from a video. So it has 2 types of inputs that we can use. And what's great is that the audio determines the emotion on the face. So a few little things, just a few pieces that the model is 3D in its nature because we are delivering it through phones and through desktops. We print it out, obviously, in the 2D space. We're not necessarily going to go into the 3D space, but the model in its nature regenerates the full face for human in 3 dimensions. As I say, we plugged it in already into a user interface, which we are thinking through the next generation of what the user experience needs to look like. You can see my face. The one problem with all of this is that it is my face. So a lot of our test stuff is actually with my face. I'm trying to push the team to use their own faces, but they're a bit shy, I would say. So this is the video that's used with me and the purpose of this is to show really how you can use different types of voices to trigger different kinds of...

I looked at the data you provided Everything seems to be in order. And the logic follows the standard protocols we established last quarter. It's a very sensible approach. Why is the bottom line highlighted in red? Are you kidding me? Did you delete the entire contingency without asking? I cannot believe we are back here again. How many times do I have to explain that we do not touch the reserve. Get out. Seriously just leave.

Again, part of the takeaways from this is it's -- the whole face is speaking, the eyes are speaking. I'm blinking. Like you could see that I'm not even controlling my eyes that well in that video. My mouth opens very expressively. My face moves a little bit. We're very excited about it. It puts a lot of pressure on getting the voice right as well, which is also something that we are learning. So we're also looking into voice partnerships and stuff that where the motion can really come across in the voice. Again, team very, very excited about this. I had to put this in just because I want to show this is -- again, this is a bit longer term in the sense that we want to -- thinking through role playing education in hospital settings, et cetera. we see a future where you go closer to the whole human body. It's not an immediate priority for us, but longer term. Having the face as well defined as the previous video that I just showed you in sync with a body that's telling the same story with the same emotion is a research challenge. We are working on it. But we have to kind of slow and steady get there starting with the face. Yes, this is, by the way, Kevin, from our business development team. He doesn't love this video, as you can imagine, but he accepted that I use a different video than my own for this. Okay. And I just wanted -- one other piece that the team always remind me, and I remind myself is -- we're going to start somewhere with this new model. We're going to bring it out. We're going to launch it to the market, but it is the V1, and we have many ideas on how to make it even better. So it will be getting something out quick, early, new, exciting that our existing customers are asking us for, getting some launch partners working with them for this and then building it out from there. We wanted to keep this very much focused on the digital human platform, but I do want to stress one piece is that we are the luckiest digital human -- we're, in many ways, the luckiest platform company out there because we have -- let's say, B2B platform company out there in the AI space because we have an internal customer. Our B2C division, they are building products. They use the digital humans -- their customers are using the digital humans every day. From the 30,000 digital humans that I showed you that have been created, 4,000 of those are, let's say, right now active and being used by our B2C division. Their customers are speaking to them globally in all different languages. I think here you see they've got 21 languages in 36 territories, and they're the first point of feedback. So they help us evolve our product. We help each other evolve the product, test the product, put it out to the field first in a semi-safe zone. And it's a zone that's directly monetizing digital humans. So just to quickly wrap up and then Sytze, I'll pass over to you. The market is growing. The market is still early. It's growing, and we are taking decisions to position ourselves to capitalize on that growing market. We are very happy about streaming avatar technology, proprietary technology that we have that is opening up lots of opportunities for us and unblocking opportunities for us and bringing back previously lost opportunities. We are extremely excited about this Unith state-of-the-art synthesis model. We've done it. We know how to do it. Now it's a case of packaging it and putting it into production. The ISO update that we've mentioned helps us in tremendous ways as we work with larger organizations, which is where we want to be focusing in the future. And then our B2C division continues to make the progress that it's doing. I don't want to be too excited. We've got a healthy pipeline. We have a couple of very interesting opportunities that we are in the final phases of. We are preferred vendors for relatively -- for very large organizations, strategic organizations, all of them also playing on this role playing training, role playing enablement use cases. And then we need to take all of this learning and we need to market it in the right way. And I think this is one of our weaknesses and areas that we have -- it's one of our weaknesses, but almost a weakness that is addressable and something that we can address with some of the right investments on a marketing perspective, both in the team makeup as well as in the investment, literally the investment that we put there. And then finally, as I say, the most exciting part is where we're going. I believe incredibly much in the direction we're taking from a product standpoint. And I think it's extremely exciting. And I think it's what's going to happen in the sense that the market -- the businesses will expect to have digital humans everywhere, and we need to keep moving in that direction. I stop there. I maybe went 10 minutes longer, but I know that I wanted to hand over to Sytze as well just for a bit of an update.

Yes. And thank you to anonymous attendee who posted a question. I think that it's a nice starting point and a nice bridge for me. It says the technology appears to be positioning Unith at the forefront of the sector. So Rakan, I think you've done a good job because this question came in at 09:42, which means at the end of the presentation. And I think clearly, your bit, Rakan, demonstrated where we are with the technology when it comes to, let's call it, the digital twin, the Avatar side. But then the real question comes, how is the company progressing on strengthening the commercial side of the business, particularly enterprise sales? Very, very correct comment because it is the large enterprises that have the muscle and also the financial ability to change their business models, partnerships, you mentioned some Rakan and account management capabilities. And I think an answer to that, one of the things we've been working on quite intensively in the last few months is looking at either partnerships, but we actually went beyond partnerships, and we're looking at more M&A type targets that actually complement what we have today because if you're really into the -- because we are in the conversational artificial intelligence world. And AI is all about data. So data is key to any AI opportunity. And if you want to interact with AI, you actually need to tap into the existing ecosystem that large enterprises have. So large enterprises can have Microsoft 365, whatever, they can have Amazon, they can use Salesforce, but the whole ecosystem of software systems that they currently use. And many large enterprises have 20-plus different software systems that in total isolation are used in their business, and we need humans to connect these systems. But if we want to have very effective communication, and it doesn't matter, Rakan mentioned consumer, so sales interaction, it can be any type of interaction. But you have to tap into the systems because you have to make that data, you have to unlock the data from the software systems. So we're looking at acquisition partners or of large M&A partners that we can buy. that actually bring this brain, Rakan called it earlier, to bring the brain to our business because the question -- that's one of the other questions that is, can you see momentum with governments and corporate enterprises. Governments are all about sovereign. They are very concerned about the data being somewhere in a large language model that we think is sovereign, but maybe it's not. So you want to have sovereign communication. That is not a signal that is not WhatsApp. But if you have sort of the core of the data, you can actually really start using the interface that we have built and which I think definitely is at the forefront of the sector. Just look at one of our competitors that just recently raised USD 200 million to build what we basically have. And I think that should resonate in your ears a little bit. They are valued at $1 billion today, and they have text to video. That's not the same as interaction. That's the easy bit, you could argue. So we're looking at that with some really, really interesting candidates on the list in our M&A funnel. So it's not about client funnel. We also have an M&A funnel today. And some of them are looking very promising, which will transform our business into far more sticky. So within an enterprise client, we solve the whole problem for them and not just the communication problem. It is a data problem and making it sovereign. So hopefully -- and that's also part of the raise doing M&A is also costing money, of course. We have to invest a bit in it. So part of the AUD 2 million is flowing in that direction as well to transform the business and really use all the technology that Rakan just explained into enterprises where the biggest opportunity for us lies in creating annual recurring revenue basically. So I think I wanted to share that. We are in advanced stages in sort of making our M&A funnel sort of with 2, 3 candidates that we think will really transform our business.

Thanks. Just conscious of stepping in because we're conscious of time. We have 5 minutes left, so probably time for 1 or 2 more questions.

I was just mid-typing -- I'm trying to use this Q&A thing to mid-type. There was a question from an anonymous attendee. How do we charge existing customers that ask us to develop more attributes into their TH, which I presume means talking head. The answer, of course, with this it depends exactly what they're looking for. We don't -- so just to be clear, we do not do any type of custom development. There are certain types of custom development that is basically feedback that goes back to our research team and would be thought through more in the context of enhancing the overall experience to all of our customers. Typically, the types of requests we get from customers are to help them improve the conversational aspect of their digital humans, connect their digital humans to other systems, embed their digital humans in custom applications, et cetera. And for each of those, we've got, let's say, pricing, we've got services work packages where each type of that has a range. And then the range often depends on the system that is not unit, right? So if we're integrating with -- into an iOS app, it's different than if we're putting it on a website. If we are writing to Salesforce, we're connecting to Salesforce CRM, it's different than just connecting to a Google sheet or an Excel sheet, which is where they may manage their customer data. So we scope out the work and then we provide a quote to them. I'm just looking through any other questions that we've got. There's many, sorry. I don't know, Gaby, if you want to try and take one?

There's one here about the disconnect between your technological progress and current market valuation. What milestones do you believe will be most important, sorry, in demonstrating commercial scalability to the market?

I can take that, but I don't know whether Sytze, you want to put a perspective on that as well.

Yes. I think it combines a bit. Scott also asked the question says you seem to be here. So I think your presentation came across. Yes, it is modern technology and anything to do with AI. It's moving very, very fast. And how do you sell it? And where is your tipping point. The tipping point is in sort of a launching large customer. And I know you have in your sales funnel, you have a few larger opportunities, very, very well advanced actually. So once you have one of the larger enterprise type clients, then that's where a tipping point sits. You need a launching customer, a large launching customer and not -- and to Scott's question, we're not relying on small and medium-sized businesses. The small and medium-sized businesses, a, they don't have the cash. To actually -- and you need a lot of them to sort of support the business. So you're looking at enterprise clients and to get access to enterprise clients at a different level that is not through social media. That is not -- that means actually C-level interaction with that enterprise client, and that goes even beyond the technology team or the IT team. It goes at Board level. It goes at C level where people have to be willing to say, how do we embrace AI because many people see it as a silo, they call it AI. And we are a one solution within that AI opportunity to replace humans with a computer, which artificial intelligence is. So tipping points, you want to answer it. One large account, one large sales that land will be a real sort of a demonstrator where you can keep deepening that vertical. So if that vertical happens to be and just for argument's sake, is in training and education, we get a very large platform supplier of training and education platforms and they start embracing our technology, then you can deepen that vertical because then you have the launching clients.

Okay. Thank you. Look, we're right on time now. So I need to leave it there and close the session today. Thank you to everyone for joining and for those who submitted questions. For those questions that we weren't unable to -- we were unable to answer today, we'll get back to you after the briefing. Thank you to everyone for joining this evening, and have a lovely evening. Thank you.

Thank you, everyone.

Thanks, everyone.

Thank you.