Conference Transcript - AIMD 2026-06-17

Craig, Analyst — Host

Coming up next is INOS, ticker AIMD, on the NASDAQ, presenting today, Jack Lew, head of corporate development. There you are, Jack. How are you?

Jack Lu, Other

Good day. Doing well, Craig. Good to see you again.

Craig, Analyst — Host

Good to see you again. Good to see you again. I suppose you'll be using your presentation today. Yes, I will. Let's go ahead and display it while I get some preliminaries out of the way. Once again, we love your questions. Submit them by pressing the Q&A button at the bottom of your Zoom window and typing into the text box. We'll keep your lines muted throughout the presentation. Here is the safe harbor statement. This segment may contain forward-looking statements within the meaning of the Private Securities Litigation Reform Act of 1995, all statements pertaining to future financial and or operating results, along with other statements about the future expectations, beliefs, goals, plans, or prospects expressed by management constitute forward-looking statements. Any statements that are not historical facts should also be considered forward-looking statements. And of course, forward-looking statements involve risks and uncertainties. Jack, please go right ahead.

Jack Lu, Other

All right. Good morning, everyone. My name is Jack Lu. I'm head of Corporate Development at INOS, stock symbol AIMD. Thank you for joining us today. INOS, we are building what we believe is an important missing layer of the physical AI, that is the sense of smell. Today's AI can read, can see, can hear, and can reason. but ai still struggles to understand the chemical information that surrounds us every day so our mission is pretty simple we're teaching ai to smell by digitizing scent and then transforming scent into machine readable data so today i'll explain why we believe smell can become a new AI data layer, how our AI node system works, and how we are commercializing the platform across semiconductor manufacturing, healthcare infrastructure, and other real-world environments. All right, so let's dive right into it. One minute. Okay, so this is our own version of disclosure statement just gonna leave here for there for a while so with that covered let's start with a simple question why does smell matters for ai so every major ai breakthrough started with a new data layer large language models needed text computer vision needed images voice ai needed audio so once ai could understand these formats entirely new industries emerged but there's always one major source of information remains largely inaccessible to ai today that is smell we often explain the opportunity using the computer vision analogy so if you recall camera existed for decades before computer vision emerged and that breakthrough really happened when the images get digitized once the images become data ai could learn from them and that shift created all kinds of new applications that are so embedded into our everyday life today ranging from having a webinar like we're doing right now watching youtube videos watching mba finals on your phones, self-driving cars, to medical imaging. We believe scent may be approaching a similar moment. The world constantly generates scent information. Machines so far simply don't know how to interpret it yet. Equipment failures, gas leaks, overheating cables, environmental environmental changes, health conditions, many of these signals appear first in the air. So the question becomes, how do you turn scent into data? And that's exactly what AI Knows our product is designed to do. So our product converts scent into machine-readable intelligence in four main steps. Our hardware captures the scent signals. we convert those sent signals into smell id data our ai model which we call the smell language model learns from that data and then we deploy that intelligence into real world environments so the important part isn't just the hardware the key point is the infrastructure just as ai requires computing infrastructure and data infrastructure we believe smell ai will require scent infrastructure i know serves as the foundation of that infrastructure we transform real world sense signals into machine readable data that ai system can understand and use and that brings us to the business model behind the platform so this slide really is the heart of our business model every deployment generates smell id data and that data improves our smell language model a better model creates more customer value more values drives more deployment and then the cycle repeats and our revenue grows so unlike text or images send data cannot simply be scrapped from the internet someone has to collect it in the real world that's why we believe early deployment creates an important advantage to us over time our smell id data set can become a very meaningful competitive mode and our customer stickiness will build so the technology matters but the data mode matters even more so the next question becomes where do we deploy first so this is the ai node system the device combines multi-sensor detections cloud connectivity ai analytics in a very compact deployment ready device we use sensor chips from leading companies so making them important ecosystem partners of ours we focus on integration calibration signal processing send digitization and ai interpretations the system is engineered for fully automated clean controlled continuous sensing we optimize for factors such as temperature humidity air flow environmental stability which are critical for reliable operations in the industrial environments the platform is also designed to be very sensitive in many applications we can detect compounds down to parts per billion levels enabling early detection of very subtle environmental changes that could signal equipment problems contamination risk or other operational anomalies device very small roughly size of a smartphone easy to deploy our goal is that our customers can place multiple AI Nodes systems throughout their facility, creating a so-called smell map that helps them to detect and locate potential problems earlier. So unlike the traditional sensing systems, our product is trainable so that our customers can continuously teach the platform to learn new scent patterns and environmental conditions. In short, the AI Nodes connects the physical environments to our smell ai infrastructure building the technology is only part of the stories the next question is where we apply it so roughly 13 years ago our team began developing ai nodes for healthcare and medical device applications that's one of the most demanding environments for sensing technologies where really accuracy reliability validations are all key and that journey helped us build the foundation of our smell ai platform including the id data sets the smell language model and the expertise required to digitize them in the real world so today we are applying that same platform across semiconductor manufacturing healthcare infrastructure and other robotics and other industrial environments different industries the same one smell ai platform so healthcare gave us the foundation and then commercialization gave us the directions and that led us to semiconductor manufacturing our first major industrial market so where do we start we start where scent already matters that's manufacturing and we start within chip factories. Chip factories use hundreds of specialty chemicals and gases every day. Tiny little changes in the environment can create big operational consequences. So the customers, these fab companies, chip makers, they already invest heavily into monitoring and process control, but we always believe sent intelligence can become another valuable layer of their environmental information. So chip manufacturing also operates 24 hours a day. So that creates an environment for continuous real-world deployment. From our perspective, it's a great place to begin commercialization. And that's exactly where we secured our first commercial order. This is where the story moves from lab to executions. So our first customer is one of the world's leading semiconductor packaging and testing companies together we are now establishing a multi-phase deployment roadmap the initial phase includes roughly 1400 ai node systems under three year subscription structures over time the roadmap would potentially lead to scale significantly beyond initial deployment as we continue to execute this is a recurring deployment model so every deployment unit generates revenues at the same time every deployment expands our footprint and that's why we view this deployment as a very important milestone this first commercial contract validates the opportunities so our next step is ecosystem expansion um we first we are first building our partner networks in Asia because Asia accounts for roughly 70 percent that is 70% of global semiconductor manufacturing capacity. So our ecosystem partners each play a unique role. Some help us expand customer base and market reach. TopCode here is a good example. They expand our commercial reach through its customer relationship and industry network. Together we are exploring opportunities across multiple industrial environments, including the front end semiconductor fabs, so these projects will help us grow our revenues and expand the network and our ecosystem doesn't really stop here since last year we have continuously expanded strategic partnerships across industries so this is one of our other partners truthful they help us reach more customers they strengthen our access within the front end fab infrastructure environments and the partnership expands our presence in the customers, their expertise in infrastructures, they are basically the backbone of these semifibes. So their expertise help us accelerate validation and deployment efforts. So together with our partners, we demonstrate how we are building an ecosystem rather than pursuing isolated pilots. So healthcare builds the foundation, the chip apps commercializing the products. And then the next step is the network and use cases. So one of our health, another use case is healthcare infrastructure. And healthcare infrastructure represents another important deployment environment. Strategically, this slide is about something bigger. it's about building a small intelligence network so every deploy new environment expand the range of real world same patterns that our platforms can understand chip apps contribute one type of environmental intelligence healthcare contributes another they all share similar things they are all very sensitive to small changes in the environment that could be crucial So over time, we believe these connected deployment networks can form a growing smell intelligence network. We call that the smell map across industries. We can now deploy through programs. We are within this healthcare infrastructure environment. We are now deploying through programs with two leading Taiwanese medical centers. They will help us support hospital infrastructure, critical facility operations. The programs are expected to generate more than 2,500 hours of environmental sent data so that we can continue to improve our models. At the same time, this platform collects no images, audios, or personal information. So again, different environment, one platform, expanded networks and that leads us to the final investment thesis so let me leave you with four key points number one we believe smell represents a major missing data layer for physical ai number two we spent more than 13 years building the technologies the smell ids and the expertise needed to digitize scent number three this year 2026 marks the beginning of commercialization through chip manufacturing deployments and the growing ecosystem of partners and also our latest new deployments within healthcare infrastructure lastly we're building more than the product we are actually building an ai infrastructure based on smell ai and we are expanding expanding that smell intelligence network across real-world environment. So as the deployment grows, our revenue will grow, the network will expand, the platform will become more valuable, and we believe that combination will position INOS at the forefront of a new category of AI perceptions. So our goal is very simple. We want to build the smell infrastructure for physical AI. We discussed the opportunity of the platforms and our strategy. To bring it all together, I'd like to leave you a short video on our mission for Smell AI and the future of physical AI.

Operator

As I mentioned AI perception, smell, AI has begun to understand how the world works from text and images to sound. the next step is to bring this understanding into the physical world from generation and simulation to perception judgment and action robots can already see and hear but across everyday environments there are still critical signals that images and sound not capture inos is building a new dimension of perception smell is no longer just a human sense it is becoming environmental data that machines can read learn from and interpret ai knows transform scent clues from different environments into information that can be analyzed and applied abnormal odors from overheated wires changes in the freshness of meat and seafood pesticide residue risks in fruits and vegetables residual chemical orders after cleaning and potential harmful odors in high-risk environments these subtle odor changes are often early signals before an incident occurs before risks escalate ai knows can detect abnormal changes early identify critical signs provide real-time alerts turning invisible risks into sense signals that can be interpreted cracked and responded to smell will become the next ai data reshaping how machines perceive the world how industries understand environments and how they make decision i knows smell your future nasdaq aimd okay that wraps up my presentation Thank you for your time today.

Jack Lu, Other

We appreciate your interest in the company. We look forward to continue to update you our progress. If you're interested, follow our social media. Thank you.

Craig, Analyst — Host

Appreciate it, Jack. Jack, a lot of investors, of course, would be familiar with AI as text, sound, images, things like that.

Jack Lu, Other

Nose and smell may be a little bit different for them. what is that aha moment that they have that really hooks them yeah that's a great question so the aha moment usually comes when people realize this that some really real-life events happened in the air before they become visible I give you a classic example I think that was about March there was a incident at Newark Airport, where the control towers, they needed a temporary evacuations because the personnel reported that there's a burning smell within that tower. And that smell was really the first indication that something went wrong. And what went wrong? Their cables overheated and started to smell weird. And this kind of events are something that most people can relate to. a gas leak, an overheating cable, a process issue in the factory often creates a smell signal before you can see it or before machines can see it.

Craig, Analyst — Host

So once our customers understand that these signals can be digitized into data, they really start viewing smell as another important source of intelligence that they can deploy with ai so that they can better understand the physical world excellent jack thank you very much that sums up a lot of the kind of questions we've been getting about that and uh just to remind everyone if you want to submit your question press the q a button at the bottom of your zoom window a text box will appear and you can then submit your question as We can only take your written-in questions today, audience. We ask that you not use the raise hand button. Another question we've got here, Jack. One of the more interesting aspects of your story is the concept of a smell ID and a smell language model.

Jack Lu, Other

Over time, do you see the greatest value residing in the hardware, the software, or the proprietary sent data being accumulated across deployments? yeah i think all three layers matters uh if you think about our ai knows platform there are three layers the hardware layer that's the white box i show you the ai layer that's the smell language model and the data layer that's being collected in the real world and you can't really get that from the internet um over the long run i think the data layer becomes increasingly important the hardware really is the gateway for the signals to enter the digital world the software creates the intelligence you can download video images text data from the internet but you can't scrape send data you gotta gotta just gotta smell it from the real world so that's why we're focused on building that small ID data sets through deployments. Just like a lot of the AOM these days, the algorithm is the backbone, but the real value is the data that feed into it and the outputted intelligence. And I think we're on the similar trajectory.

Craig, Analyst — Host

Jack, you've chosen semiconductor manufacturing as an early commercial focus. What makes that industry particularly well-suited to demonstrate the value of AI-powered scent intelligence?

Jack Lu, Other

Yeah. So for us to build our platforms, we need a lot of data that gets churned out 24-7 so that we can optimize it and create that data flywheel. And chip factory is actually one of the ideal environment for that because they are highly chemical intensive. and small changes can have meaningful operational consequences and these chip makers they already invest heavily in the monitoring and processing they want to control risk before it escalates because they simply cannot affect for downtime we believe sent in sent intelligence can become another valuable layer for them because they already collected all kinds of data vision data pressure data even vibration data a lot of these fabs are now fully automated so having that extra layer of information can help them discover the unknowns and they are 24 7 environments so they are really a perfect place for our product right now we are working with the biggest back-end part of the supply chains and through our partnerships we are now expanding into the front-end part of that cheap supply chain which is the one of the highest valuable part Jack are you finding that the core technology transfers well across industries or does each new vertical create its own unique opportunity to expand the platform yeah as you can see from my earlier slides the platform really transfers well right now we are what we are deploying in the chip factories and in the healthcare in the hospitals are actually the same product so the changes in the application and the type of environmental intelligence that's the change but the concept remains the same so for us expanding deployments help us expand what the platform can understand just like chat gbt you start with a simple chat box you feed it with all kinds of information and then become domain experts and then you can do all this agentic work as they are doing today now jack how do you envision olfactory technologies change olfactory intelligence sorry changing the capability of autonomous robots over time yeah so you know we the humans we don't rely on visions alone right eventually we think robots probably won't either I think smell can become a very important sensor input that helps robots to better understand their surroundings, detecting things that vision cannot do. Just like people can smell smoke or gas before they see the source, robots may eventually use scent intelligence to detect environmental changes. and actually right now we already have some projects with some of the robotic customers in asia to add that extra sense of smell onto their robotic platforms jack your healthcare heritage your very ticker is aimd what advantages do they provide in developing a platform capable of addressing such a broad set of markets yeah medical is really our root that's where everything started for us so we spent about roughly 13 years developing and refining ai nose technology in a very divided environment because accuracy reliability and validation matters for medical devices. So that really just leveled up our entire platforms. And now that foundation is now helping us to expand into chip factories, healthcare infrastructures, and other applications because the concept is the same. They demand accurate, reliable, scalable sensing based on smell. final question jack we've got about 30 seconds for you to sneak this one in um how do you think about the eventual size of the smell data economy if digitized scent becomes as commonplace as image and voice data is now yeah i think it's very early to put an exact number on that opportunity but conceptually I would say computer vision 30 years ago versus now it's a lot it gets a lot bigger so we've seen this movie happen before I mean I just saw a data right video now accounts for roughly 75 percent of the global mobile data traffic so people always take photos videos so So what's changed is they become machine readable data and you can send them around and process People have always smelled things too. So if send can become a data, it could enable entirely new category of AI applications like what we are driving today. So our focus today is to build the infrastructure and data sets that could help to make that future possible. So I would say we're at that beginning of that infection point and what you're seeing with computer vision now, that's where we eventually will become with smell.

Craig, Analyst — Host

Jack Lew, thank you very much. It's INOS Incorporated, AIMD on the NASDAQ. If you'd like more information about INOS, write us at aimd at redchip.com or call us at 1-800-REDSHIP. Jack, great to see you again. Thanks for the excellent presentation.

Jack Lu, Other

Thank you. Bye-bye.