Video: One Unstructured Data Foundation for Your People and AI | Duration: 3364s | Summary: One Unstructured Data Foundation for Your People and AI | Chapters: Welcome and Introduction (6.16s), Industry Challenges Converge (247.935s), Company Updates (379.37s), Data Journey Framework (483.555s), Platform Overview (658.95s), AI Stack Integration (930.475s), PSYCHIC Framework (1002.825s), DriveIQ Platform (1402.065s), AI Activate Service (1684.655s), ROI and Productivity (1944.165s), Pilot to Production (2042.87s), AI Activate Demo (2126.065s), Enterprise Search (2389.66s), Resilio Integration (2448.29s), Business Outcomes (2687.535s), Ecosystem Integration (2902.65s), AI Governance & Risk (2989.74s), Product Selection Strategy (3094.86s), Copilot Integration (3180.32s)
Transcript for "One Unstructured Data Foundation for Your People and AI":
Alright. Good morning, good afternoon, good evening to everyone. I'm Nick Burling, chief product officer here at Nasuni. Thank you so much for spending some time with us today. I'm really excited to share a bit about what we've been up to and some of the key new products and capabilities that we've announced, just a few weeks ago. I am glad we're being joined here by Dave Castagnola, our CRO. And, yeah, really looking forward to taking you through it. One of the key things that we found is it's really helpful to give everyone context in terms of what are real world scenarios for how our capabilities come to life. So with that, we're actually gonna get ourselves kicked off with a little bit of a video. 10/06/2026. I was a design associate back then, working on my first big project. That was the moment I saved the first design for the new Unify Bridge. Across town, crews were taking the old bridge apart. Our new one existed only as a file. From that first save, it was protected and understood. What it was, where it belonged, who could access it, and how it should be governed. It joined every drawing, contract, spreadsheet, map, and site photo connected to the project, along with millions of files across the enterprise. Teams in London, Singapore, and New York could work from the same source. And with high speed edge access, even Jim could open our largest design files right from the job site, all without stepping on each other's changes. And there were plenty of changes. Every save became another protected version, building a clear history of how we got here. Finding something stopped meaning hunting through folders. We could find the right content in seconds with the same permissions already in place. Then one day, I opened an invoice email that looked completely normal. It wasn't. The attack moved fast, but it couldn't erase our work. The affected files were contained, and we quickly recovered clean versions with the entire project history still intact. Then AI became part of our team. I could ask, have we solved something like this before? Or what's the total personnel spend to date for the project? Instead of finding, opening, and comparing everything myself, AI could work with the live project data I was allowed to access and the knowledge around it. I got to answers faster with the context to make better decisions. Soon, the bridge was no longer just a model on my screen. I could stand at the job site and watch it become real. A few years after that first save, we cut the ribbon. The Unify Bridge was live. Unify Bridge changed a lot of things, including my career. I'm a director now, leading a large team. But Unify Bridge is still one of those projects I come back to. And Unifibridge kept working too. Its files, its history, and everything we'd learned were still there. Years later, one of my teams sent me a proposal for a new suspension bridge, asking if I'd worked on anything similar. Brought me right back to the Unifi days. I asked AI, pull up Unifi Bridge. Show me the designs, the decisions, what changed, the problems we hit, and how we solved them. Years of work came back in seconds, Not just the files, but what we'd learned. Start there, I told her. One file became a project. The project became knowledge. And that knowledge became the starting point for what came next. That's the life of a file and a designer, both backed by Nasuni. Alright. I don't know about you, Dave. I never got tired of watching that thing. Nick is fantastic. And, good morning, good afternoon, good evening, everybody. Dave Castagnola here. Sandra got a bridge. She got a promotion, and let's be honest. She got the version of the story where the data cooperated. If Sandra, gets. to look back years later, and most of us are looking at a budget meeting next quarter. In fact, I had my budget meeting last night. So let's talk about what's happening in the real world right now. Three forces are colliding in the industry, and and they feed off of each other. First, we've all lived with for years rationalizing and optimizing, do more with less, cut complexity and cost while increasing productivity. And that pressure never goes away. Second, AI has amplified it. Companies are investing in AI like never before, and that spin has to come from somewhere. So it puts even more pressure on that first priority. And third, and this one is specific to our world, the AI boom has created a chip shortage and is driving up hardware storage costs, availability, and in some cases, three to x to four x cost increases. This is not a supply chain hiccup that we've all lived through a few years ago. It's a structural change, and it's gonna be with us for the next few years. So you're being asked to fund AI while the storage underneath it gets more expensive while still being told to do more with less. And every one of those makes the others harder. And that's why our customers are rethinking their modern data infrastructure. And you can see it across the whole organization. The c suite says we're investing in AI, but we can't get the data foundation right. The IT operator is saying, we're spending all of our time keeping the lights on, and the end user is saying the AI doesn't have what it needs. Different people, different seats, same root causes, data readiness. And it's the number one blocker that we hear today, from the market. Okay. And that's the problem we spent working on this year in 2026. So I just wanna show you real quickly what's changed at Nasuni. I won't talk through the whole timeline. Just three things. One, we're proud of our customers and their success, and we listen to them. We learn from them. That has been part of the DNA of this company since it started. We have an annual survey. And, this year, we surveyed a thousand decision makers. 97% are deploying or piloting AI agents, but only 18% are at scale. And we're hearing that gap between pilot and production is the data foundation problem in a number. Two, we had two fantastic companies join Nasuni this year required to to gain more capability. Resilient gives distributed teams local speed, file access with no VPN. DriveIQ, our most recent acquisition, brings content intelligence and governance across 40 plus repositories so you know what data you have before you point AI at it. And then three, we built. AI activate gives governed AI access to your file data through MCP with permissions enforced through the query. An enterprise search finds anything across your state in seconds, again, permission aware. And we're proud that for the sixth straight year, the North Face scoreboard award with 98% customer satisfaction and an NPS of of 88. The new Nasuni can meet you wherever you are on your journey, whether you're clearing up the state, speeding up your teams, or putting AI to work on your data. And we're really pleased that you joined us today. Thank you so much for making the time. To show you how it all comes together, let me throw it back over to our chief product officer, Nick Burling. Nick, take it away. Alright, Dave. Thanks for, for setting that context. A couple little housekeeping things that I should have covered at the top. Number one, please do take advantage of the q and a. We will definitely have some time at the end to address some of the questions. Anything that we can't get to, live here during the session, we will certainly follow-up with you over over email. And we are gonna be running a a quick poll, towards the end there, so please do participate in that. It's really helpful in terms of how we can best follow-up with you based off of, of the content that we covered today. So like Dave said, you know, one of the key things is obviously customers are at various different points in in their journey, in this in this world of thinking about unstructured data and and in particular, how can you leverage that as a key asset around your AI strategies. And in particular, one of the acquisitions that Dave men... May, mentioned, DriveIQ, has really added on a step in this journey and in this process that was really pivotal to us in terms of looking at the portfolio and the platform that we've built out and filling in a gap. And that's all around this idea of understanding the data that you have. Over the last fifteen plus years, you know, Nasuni has built up a phenomenal solution from a manage and protect perspective. And then we've also acquired as well as built out organically key capabilities around this idea of activating that data for teams and for AI. But so we think about understand. We know the reality is even for the customers that we have today, many times they do have data silos, whether those are in SaaS platforms like SharePoint OneDrive, Dropbox, Google Drive, etcetera, whether there are remaining, on prem data silos. The reality is getting your arms around that data and understanding not just where the data resides, but what is in the data has never been more important than in this world of AI. Because if you don't understand what the data contains, then you don't understand the risk profile associated with that data or the value that it could offer in terms of thinking about activating that data for new AI solutions that you're building out. And then manage protect really are those foundational steps. Right? You wanna be able to unify that data, have a single global namespace. You wanna be able to ensure that you can protect that data both from a disaster recovery and backup perspective, but also from more advanced types of cyberattacks such as ransomware or new types of infiltration that focus more around data exfiltration. And then last but not least is the Activase piece. Right? As I talked about, Resilio plays a key piece here as well as the AI activate solution and enterprise search that I'm gonna cover as we go forward. So, hopefully, this gives you a little bit of a sense of kind of that journey that we'll be talking through. And now let's dig in a little bit deeper in terms of the new product announcements that we've made over the last few weeks, and then I'm gonna go into each a little bit more and talk more about the value that they can provide to all of you and your organizations. DriveIQ, as Dave talked about, this is all about content intelligence. It's about being able to go beyond just the metadata and down into the contents of the files themselves. The fact that it can extend beyond... Certainly beyond just the data that's on the Nasuni platform, Dave mentioned 40 plus connectors. And so that breadth of coverage as well as the depth is what make this a fundamental part of that understand phase. And that's gonna be everything from thinking about optimizing your data, identifying redundant, obsolete, or trivial data, but as well as being able to define and implement and enforce governance policies as you start thinking about activating that data for AI, but in a secure and governed way. AI activate. This is a a a Nasuni built solution that we just launched and went GA at the end of September. And this is all about providing standards based access for optimal performance with your AI solutions to all of the data that lives in the Nasuni platform. Standard is all about the MCP. Yeah. You'll hear me talk throughout the session today about Nasuni as being a platform, and we we take that very seriously. And part of being a platform is ensuring that whether your end users are using Copilot or ChatGPT, Gemini, Cloud today, whether you're building agentic solutions in Azure Foundry or whether they're in AWS Bedrock, when you make changes to those AI solutions and workflows, the underlying platform should not have to change as well or block your ability to implement best of breed solutions from an AI perspective. Another key part of AI activate that I'll talk about is the secure permissions enforcement as part of it. Enterprise search, pretty self explanatory, but nonetheless very important. You know, we have... We're customers that literally have billions of files, and they need to be able to search across all of the volumes of data to be able to find the field with the right haystacks in it and then be able to dig within each haystack to find that that proverbial needle. And then last and definitely not least is Resilio. As Dave mentioned, this is one of the first acquisition we did earlier this year and really provides a tremendously complementary solution to Nasuni's hub and spoke architecture, extending our edge out with ultra high speed access without the need for VPN. So if you are a civil engineering firm with engineers out in the field working on big design files, if you're in energy doing exploration out in the ocean and need to get geospatial data off of a vessel, this really is a tremendous platform from a data orchestration and access perspective. So that's a high level view. Let's start to drill in on these in a bit. And to do that, I wanna frame it up in terms of thinking about how this platform comes to life. And so think about really four key pillars. Underneath is the core. Right? That's the foundation. That's what allows us to unify and protect all of your unstructured data in that single global name space. It truly is hybrid in the sense that, most of our customers choose to store their data in one of the major object storage providers, whether that's Azure, AWS, Google, Oracle, but, also being able to support on prem, object storage as well. The other hybrid component is is our intelligent caching at the edge. So being able to make sure that whether they're users or applications or agents and AI solutions can access that data wherever they are, literally anywhere in the world. Next, you have cyber resilience and governance. And so we think about, you know, I talked about ransomware. Obviously, it is still a pervasive threat, and we have advanced capabilities around detection, prevent mitigation, and then reporting, and most importantly, rapid recovery. But we also continue to expand that out with some of the governance capabilities that DriveIQ brings us and advanced cybersecurity capabilities that we're building into the platform. The The next pillar is data intelligence. So we talked about the importance of being able to actually see into the files, not just the metadata, actually be able to understand, do entity extraction, do labeling, tagging to be able to define policies and action sets as part of the governance story. And then also to have operational visibility across the entire portfolio, that's our ops IQ solution. And last but not least is data activation for teams and AI. This is where AI activate sets. This is where search sets, and then Resilio in terms of activating that data for for your individuals, for your teams, for your end users. Alright. So let's dig in a little bit more. The other key piece here is thinking about how Nasuni sits in your overall AI stack. So this is a very simplified diagram, but the key point here is think about these major layers. And sometimes we see prospects and customers that are doing a lot of work with some of the phenomenal solutions like Databricks and Snowflake, Microsoft Fabric, as examples. And too often we see, sort of an assumption of, okay. To make this data activated for AI, I need to make copies of it into my data lake. And the key thing to think about is that Nasuni really sits as one of those foundational layers from an unstructured data perspective. And the reality is, you know, depending on who you talk to, that can be anywhere from 80 to 90 plus percent of all of your data that lives in that layer. So we sit right alongside that, that structured layer and then provide that standards based integration. So whether you're building out any of these AI solutions or whether you just have individuals wanting to interact with that core institutional knowledge that lives in the operational file layer, they can do that, in a secure, optimized fashion. Now one of the other key things is when we go out and talk with customers and really over the last twelve months, we talk with analysts, we talk with partners, one of the challenges that we consistently see is how do I think about this idea of data readiness in the world of AI? And different people have kind of different ways of framing it up. But what we found was we really wanted to have a common language. And I mean common in a way that it's not designed to be exclusive to Nasuni by any stretch, but a way that you and all of your stakeholders with your organizations can go out and have a way of thinking about in a structured fashion, how do I assess our readiness for AI? So we developed PSYCHIC. Now, obviously, it's a little tongue in cheek in that it's a a memorable acronym, but the reality is each of these attributes of your data really is pivotal in terms of thinking about this idea of AI readiness. So if we go kinda top to bottom, proprietary is the starting point. And frankly, it might seem like a little bit of a no brainer, but it's really important that in this world where everybody has access to the same models, whether those are the frontier models, whether you're deploying local models to help with optimizing cost, whether you're even training your own models, at the end of the day, that's not where differentiation lives and that's not where you're gonna get competitive advantage. You're gonna get competitive advantage by being able to leverage your proprietary data. That's the goal that you have, and so it's critical that any solution that you have is gonna be able to take advantage of that proprietary data. Let me get to secure, and it's not just secure in terms of can I enforce permissions as users or agents on behalf of users are interacting with the data? This is also where that governance concept comes in. Right? Because it may be the case where you need to know, does this file that these users have access to have PII in it? And if so, can I tag it? Can I prevent AI from ever being able to see that data even if technically those users do have access to the data? Secure is a key key piece. And then also thinking about as you're starting to look at deploying of agentic solutions. You know, these are like other users or applications, but they don't need sleep, and they can operate a hell of a lot faster. So you wanna make sure you have the ability to audit, to understand what data they've interacted with, and to be able to recover quickly if there is an issue in terms of an agent going rogue. Now yours is an interesting one. Sometimes we hear when we talk to to customers and prospects, well, what's the difference between yours and proprietary? This is all about what happens with the data as you start to leverage it with AI solutions. You know, you may... I think lots of us have heard sort of nightmare cases of, you know, source code getting out and being used with some of the Frontier models, and now, technically, that data is out in the public domain. It's really critical that you have not just the visibility, but the the controls to prevent the data from ever leaving your tenant, based on policies that you define. Current, this is critical. And I talked a little bit about sort of thinking about the difference of copying unstructured data into structured data platforms like Snowflake, Databricks, etcetera. Current is all about making sure that you're leveraging the latest data in that operational file layer. This is the data that's driving your business. And as you start leveraging AI solutions with that data, you wanna make sure you're working off of the latest version. And and, again, this could be the key in terms of competitive advantage as you're starting to drive business decisions in automated ways based off of the AI and agentic solutions that you're deploying. Hybrid. I talked a little bit about this in terms of of Suny's architecture, but the reality is AI is hybrid as well. So that architecture does need to be able to support that. And so what do I mean by that? You wanna be able to take advantage of compute wherever it's optimal, whether that's from an availability perspective. So as many of you who have tried to deploy compute within different cloud regions in the hyperscalers have probably experienced, sometimes those resources are not available in the region where your data resides. And we have a saying, you know, make sure that you're making data available to your teams wherever they are and bring your AI to your data, not your data to your AI. And so a hybrid architecture, again, needs to be able to make that data available. Let's say you do some model training POCs in the cloud, but then decide for cost reasons to run those on prem in a colo or or in your own data center. Again, the underlying platform needs to make sure that that data is accessible wherever the AI services and compute are running. Indexed. This is really important. When you think about a lot of us, I'm sure many of you on on on the webinar today, as you've been leveraging AI and you understand token consumption and and the reality of that cost that can be driven up, token efficiency is a key piece. And making sure that your AI is reasoning over, an index of that data so it's not spending all of its tokens just trying to find the right data is one key piece. But then the other key piece is all around accuracy and avoiding AI slump. If you try and run, you know, Copilot, ChatGPT, Cloud against tens of thousands, hundreds of thousands, even millions of files on a given volume, without the benefit of an index. You're not gonna see good results. Last but not least is an interesting one, continuous. And so if you think about the life of a file as a great example, that bridge evolved over time. The ability to go back and look at not just did we do things like this before, but how did that progress? Can I look at the version history for all of the files associated with the project, all of the edits, all of the interactions from the different designers that were working on it? And is there intelligence that I can glean from that evolution? Another interesting area, and frankly, this is sort of very early on, is as you think about maybe you do decide to train your own models. Well, if you're just looking at the latest version of your files and your your unstructured data, that's really at the surface, but there's this iceberg that sits below that surface of all of those versions that live over time. And that could become a tremendous asset as you think about not just inference use cases, but also thinking about more advanced training use cases. So I'm gonna come back to psychic, when we wrap up here to sort of help you all see how we grade ourselves against this. And then, hopefully, you know, this is something that you can take out of the session today. And if you have interest in more of this, we've actually built a whole assessment around it, and this can start to become, a key framework for you engaging with other folks within, excuse me, within your organization. Alright. So let's drill into DriveIQ a little bit more. So Nasuni plus DriveIQ really is filling a key gap that we had within our portfolio all around that understanding and that content intelligence and governance. Now a key piece that I'll highlight again is that DriveIQ works against those 40 plus connectors across your entire estate. It's also an extensible platform where we can build out new connectors to integrate within systems within your environment. And we will continue to invest and support in leveraging DriveIQ across the entire unstructured data estate. Now that gives you that AI powered content intelligence, the ability to define those policies and enforce them through what are called action sets. And we do believe when you do couple that with leveraging all of the power of the Nasuni platform, you really get into a one plus one equals three, five, you know, whatever you want. So the better together here is tremendous. And the other piece is coming back to that understand phase. As you think about engaging with Nasuni, imagine being able to look across your entire unstructured data state, identify what data makes sense to move to a platform like Nasuni, what should be archived right out of the gate, and looks at things like maybe you are big SharePoint users and you have lots of overage that's driving cost. There's a tremendous amount of flexibility and power that you can get by providing DriveIQ just on its own, but then also as a conduit to bringing data to Nasuni and then ultimately enforcing, defining and enforcing governance once that data is on Nasuni. So if you think about some of those challenges that we hear, you know, the inactive data, duplicate data, what you can think of as as raw, the redundant, obsolete, trivial data. The unknown piece is is a key one, right, because that's risk, particularly as you start... Think about allowing individuals to leverage AI against the data that they do have access to. Unsecure over shared, certainly permissions, and many of them understand what permissions exist. They may have built up over time, but they may not be ones that you want to still be in place. And then ultimately helping with that that inaccessible or siloed data and sprawl that you see across multiple different storage silos. So now once you leverage DriveIQ, it gives you that content intelligence. It gives you the richness, the ability to remove duplicates, the ability to, again, define policies and force those policies. And that gives you not just more secure governed data, but smarter AI answers because you've got the right datasets being exposed to AI. You've got auditable results against how that data is being interacted with. And so when you think about this across the board, we're gonna talk quite a bit, as we get towards the end here around, you know, business impact, a... You know, ROI. These become critical. And we can look at this across just raw, storage optimization and underlying costs, but then also in terms of business impact. So slides are great, but we always like demos. So I'm gonna ask Cassie to go ahead and give us a quick little video on DriveIQ. As content volumes continue to surge, organizations face growing pressure to cut costs, reduce risk, and accelerate AI adoption, yet unstructured data remains their biggest blind spot. Poorly governed, poorly understood, and rich with untapped strategic value. DriveIQ's intelligent data management platform empowers organizations to take control, unlocking the full value of their content while dramatically reducing operational overhead. Your data stays continuously relevant, organized, cleansed, secure, and business ready. Engineered to meet the demands of the world's largest organizations, DriveIQ intelligently scans and analyzes your content, transforming the results into rich visualizations, interactive graphs, and dynamic dashboards that uncover deep and actionable insights. Migrate, classify, and govern your content. Enrich file metadata, protect sensitive information, and intelligently archive files to drive down storage costs no matter where it lives, on premises, in the cloud, or across hybrid ecosystems. Make your data work for you. Insights are instant, compliance is effortless, and every decision is powered by accurate information. Thanks for watching. Alright. Excellent. So now let's move on to AI activate. So So as I mentioned, this is a new service that we just rolled out at the end of September, and we're extremely thankful to the customers that worked with us as part of the preview program for this to help us, build out some case studies and and proof of value. So Charles, Douglas here from McKinnon Creed. If you... Actually, those that paid a lot of attention here, have we done this before? The reality is that life of a file video, and the whole scenario around specifically that question that Sandra asks, was really built off of this use case that Charles brought in terms of his engagement with the pro pro... The, preview program. So in this scenario, you know, acquired a new dataset, terabytes of data that they didn't have any visibility into and wanted to be able to get insights and then tie that to other project work that was going on within the firm. So when we think about AI activated, sort of three key things that I'll call out here. First and foremost, there's no new infrastructure that that you as the customer have to deploy. This is a 100% hosted service. All of the data that that we work against still lives in your control inside of object storage in your tenant. But we provide all of the infrastructure both in terms of the index as well as the MCP instance. So, again, it's simple for you to come into the Nasuni portal, define the scope of data that you want to activate for AI, and then deliver that to your end users. The next piece is security. So it's critically important, again, that AI not get access to data that it shouldn't. And so as users are working with Copilot, Chuck TpT, Cloud, etcetera, you can rest secure knowing that only files that they explicitly have access to will ever, be interacted with through the service. And I'm gonna talk a little bit more about sort of the idea of how we do that in a second. The other key piece is coming back to that best of breed selection and the fact that Nasuni is an underlying platform. So whether you're changing your your chat client or whether you're going from agentic solutions built in one one cloud to another or even on prem, the underlying platform and benefits should not change, and they don't with AI activate. So leveraging MCP makes sure that we have standards based integration, and then we've built optimization into that that I'll talk about in a little bit more detail. Just to give you a sense of kind of the overall architecture here, so from a customer perspective, your storage, again, still lives in your world. You own you own the data. And then we provide the ability for end users to then be able to interact with that leveraging their chat client or AI solution of choice or, again, agentic workflows that you're building out. There's an indexing service that indexes all of both the metadata and contents of the files within whatever scope you have defined. And you do have very fine grain control over that. There's includes and excludes in terms of paths and file types. And so you can really make sure... Because when you start talking about leveraging AI with the data, that scope and that context is really important. Again, you don't necessarily wanna have individuals trying to leverage ChatGPT, Copilot, Claude with millions and millions of files. You wanna understand what their use case is, whether that's the legal department wanting to analyze prior legal contracts, engineers as part of a project, etcetera. So the next bit is coming back to that security component that I talked talked about. And this isn't a... This is a key piece, and we talk about pre query filtering of the data. So so... Or pre query enforcement of the data, I should say. So the key piece here is, again, we maintain the security context. So if one of your users from legal department is wanting to go in, the prompt comes through. Let's say they're using Copilot or or ChatGPT, and it makes its way in against the index that is filtered based off of their permissions. It is not done post query. This can be another approach where you go do a full retrieval, get all the results, and then and then filter from that point. And that's not a secure approach because if you get a man in the middle, they could have access to that entire payload. Now business impact is really important. And, you know, as Dave talked about upfront, this is something we see across the board from the the c suite all the way down to individual contributors in IT that you gotta be able to understand what's the impact of the the spend that we're throwing against AI. So we look at it. We think across these three kind of big buckets here. Token, tokenomics is a big a big focus, certainly is here at Nasuni as well, and significant savings by being able to focus on avoiding AI doing what we call the research problem. And I don't mean research in terms of looking up information and facts, etcetera, but rather having to search over and over again to find the relevant data within the dataset. Then there's productivity. Obviously, this is one that I think everyone is aware of if you think about or anybody that's had their hands on on some of these AI solutions. It can be tremendous in terms of the impact in reducing manual and trivial work or automatable work. So looking at, you know, roughly 11 and a half full time equivalent employees per year, that's a significant productivity win. And then ROI. Now, you know, this is gonna be on a range, you know, depending on how many users, what level of activity they're driving over what type of data. But the key point is this is not a, you know, one and a half or a two x. There can be real gains by being able to leverage the value that's in that unstructured data, but in a secure, governed fashion that's optimized for enabling AI to reason over the right data instead of just trying to find it. The other thing I'll say here you can see at the bottom, if you go to the website, you'll actually be able to take, take a test drive of our ROI calculator. I highly recommend you do it, not just to get some insight in terms of sort of where your organization sits, but, again, this helps provide a framework for how you can start thinking about other stake... Stakeholders within your business. Now the other key piece is, you know, we actually, ran and, Dave, on the... One of the slides that you talked about, we do a... An annual survey of over a thousand different enterprises, for multiple industries and geographies. And one of the key things we found is that while 97% of respondents said that they are actively building out, Agentic pilots, only, I think, less than 20% of them actually have them running at any level of scale. And so when we look across that, how do we help customers go from pilot to real ROI and do that, you know, within a time frame of relevance, you know, not not asking for years for your CFO to be able to prove out value. So faster time to value is absolutely critical, but also with an eye towards lower cost and risk. So tokenomics, but also, the potential cost of it if there is a security issue, if if data is, you know, made available to services that it shouldn't be. And then ultimately, again, tying back to that ROI piece, being able to have outcomes that you can take back that aren't just, you know, individual use cases but are actually really demonstrating business impact. So, again, instead of just looking at at the slide content, let's take a look at a quick demo of AI activate. Most enterprises sit on millions of files, potentially petabytes of data. In those files are the answers their teams need. Finding and synthesizing them is the problem. Let's watch how Nasuni's AI Activate can help an organization like Northbridge, a holding company with about 3,000,000 files across the parent organization and three operating subsidiaries, a Chicago ad agency, a San Francisco architecture firm, and a Midwest industrial manufacturer. What you'll see is a user querying the Nasuni AI Activate MCP server through their enterprise ChatGPT subscription. AI Activate authenticates the user through single sign on and natively enforces their existing file level permissions. The user only has access to content they're already authorized to access in the file system. We'll present four questions a user might ask and watch ChatGPT answer them. First, an easy one. The kind of question your legal team handles weekly. Show me all Northbridge counterparties with mutual NDAs. A simple list across all four entities. The answer lives in a single tracker file somewhere in legal and compliance. If you knew where to look, you'd find it in a minute or two. AI activate finds it without needing to know. Now a little harder. Which subsidiary has the highest key personnel concentration risk? There's no folder called concentration risk. The answer requires interpretation, recognizing that one creative director leading two of four agency clients exposure. The source is a personnel register most users have never opened. Manual search won't find it, and LLM walking the file tree wouldn't know to look. Question three. Which IT vendors are used across multiple Northbridge subsidiaries and what's our consolidated risk? This is a multi file synthesis. The answers live in two separate documents, an IT vendor risk assessment and the group's preferred vendor list, both in different parts of the file system. AI activate brings them together into a single consolidated view. Last question, and it's the hardest. Which counterparty engagements have any combination of missing or expired NDAs, missing MSAs, or open compliance gaps? Gaps. This is exclusion synthesis. Finding what's missing, broken, or expired across five document types in multiple folders. Without an index, this question has no path to an answer. A team could spend days assembling it manually and still miss something. Now AI activate is reaching back into the index for additional context, pulling specifics from individual subsidiary records to support the synthesis. The model is connecting threads across the parent and all three operating companies. A contract tracker in one folder, a permit file in another, a quality record in the fourth. Now notice the prompt. ChatGPT thinks some of the underlying data may be sensitive, so it's asking for explicit confirmation before working with it. It is now assembling the full picture, what's missing, what's expired, what's in remediation across the entire group, sourced sighted and grounded in actual files. Four questions, four answers. Sourced, cited, and grounded in actual Northbridge files. No content migration, no new infrastructure, no copying your data into third party systems. Just your existing Nasuni files made AI accessible. Your files hold the answers. AI Activate makes the connection. Talk to your Nasuni team about AI Activate. Alright. Excellent. So now let's move on to enterprise search. So search, obviously, pretty standard use case. I think the key here is the scale at which you can do it and then how that ties into this broader picture of data understanding, activation for the teams that need it, and then ultimately getting it AI ready. So coming out in early December, we're rolling out our enterprise search offering. Again, leveraging a Nasuni hosted, index of all of that data, enforcing permissions so users are never searching against data that they don't have access to, and then integrated right within Nasuni's web access platform, which is part of our Nasuni edge. So, again, fully Nasuni hosted. Just go into Nasuni portal, define the scope in terms of which shares are configured for your web access solution, and now your end users have that global search across the entire state. On to Resilio. So, as Dave mentioned, we acquired Resilio earlier this year. And, again, we're grateful to to one of our customers here with the face, who talks about the impact that this can have. And and, really, I love I love the end of this quote in terms of another world of improvement. And that's typically what we see when you think about being able to leverage the power of highly optimized peer to peer protocol that can leverage all of the bandwidth made available to it, avoid the overhead of VPN, but still provide a lot of the capabilities that customers need that they've been used to with Nasuni. So when you think about this, there's a variety of different use cases. You know, one of the first ones that really drove our partnership and then ultimately the decision that the two companies really should be one is all around VPN less and hybrid work. The reality is whether you're in professional services and manufacturing, energy, media entertainment, etcetera, the reality is you have key people that are working at different locations, and many of our customers are some of the largest, most distributed organizations. It's critically important that those workers, whether they're working from home, whether they're at a remote office, or whether they're out in the field, that they have access to that data, that they can collaborate on that data and do so in a way that does not introduce conflicts or overriding with their colleagues that may be working in different locations. But some of the other key use cases may not be ones that are kind of obvious right out of the gate. One, is thinking about, overall or what we call the far edge. So I talked before about, you know, you might have, individuals that are out on vessels in the North Sea running LIDAR and scans of, the ocean bed and and need to be able to get that large dataset off of that vessel. Don't wanna have expensive physical storage there because they need to get it not just protected, but then available for others to analyze as part of an overall data processing pipeline. Another is thinking about that AI readiness piece again. So if you come back to what I said before about, you know, bring your data to your teams wherever they are, but make sure you're bringing AI to your data. Well, you may be in a scenario where you do have your data living inside of a particular cloud or region, and you wanna be able to leverage, compute running in a different location, maybe in a NeoCloud. And so Resideo can be a key value prop here in terms of thinking about data pipelines and very, very high speed availability of that data out to that compute where it is without having to create additional copies of it. So it's a it's a pretty flexible solution that really complements the Nasuni architecture, against all of these different use cases. Now to give you a little bit of a a sense in terms of kind of architectural view, so in the center here, you'll have your Nasuni volumes that again live inside of, Azure Blob Storage, inside of AWS s three and Google Cloud or on prem and Oracle, etcetera. But now you're able to think about extending out that very, very high performance, capability, whether you think about power users that are using, you know, very heavy design files like Civil three d, AutoCAD, etcetera, or whether you have those remote users. Maybe, if you're in the the media entertainment space and you've stood up a studio and need to be able to have data captured and readily available and protected at remote locations before you tear down, filming on location. So very flexible and also helps in terms of thinking about overall footprint of your Nasuni infrastructure. So instead of necessarily having to or wanting to deploy multiple Nasuni edges at different regional offices to provide that local, performance, to to your your users and workers there, you can now have a very lightweight client that lives on top of their workstation, laptop, even mobile device, providing all of that data replication and and access and even offline, access to the data as well. Alright. So I've talked a lot about the different capabilities both in terms of the two acquisitions, that we made as well as the new services that we've rolled out with AI activate in enterprise search. But at the end of the day, what you care about is business outcomes. And so, as I mentioned, coming back to the psychic framework, and so you don't think that, you know, we would just grade ourselves a 100% across the board. There's a couple of key areas that I will highlight that that we are working on continued investment. When we think about proprietary, it's not because we can't manage or or allow you to take advantage of your your proprietary data, but rather there are certain workloads that Nasuni is not a fit for specifically from a high performance compute perspective. Now that said, one of the key investments that is coming in, the first part of next year is going to be a true scale out high performance, architecture for our edge. So we do expect to close close that gap in the very near future. Then if you go down to continuous, I mentioned the potential of being able to, yes, leverage that version history. This is something that we're actively looking at today because we do think it's a unique value prop that we are positioned to provide. When you think about something like AI activate being able to access that version history. So this is something that we're partnering with a lot of our customers on right now to look at sort of what are the core use cases, and that's the key piece here as you wanna be able to leverage that that wealth of data that lives in that version history, and what could that unlock as you start to explore more and more AI use cases against that data. So then if we kinda boil it down, what are the key outcomes that you take away from today that you could go back and talk to your your colleagues, to your management, and other stakeholders? Certainly, cost is a critical piece. And, the reality is we know, all here that IT budgets are not growing at the rate at which you're being asked to deliver results. And so you've gotta be able to drive efficiencies and savings, because then you can roll those savings back into investments that you make in, you know, bottom of... I guess, better way of saying is bottom line savings that can feed top line growth and productivity. The other key thing is you've got to ensure, particularly as you're planning to leverage this this gold mine that is your unstructured data, and you start to build real business processes around that and solutions around that, we can't afford for the... That data to not be available. So So you've gotta make sure that you've got the resilience and the uptime to be able to make big bets on AI being able to leverage that operational file layer. And then, ultimately, being able to drive efficiencies in how you're leveraging AI, like we talked about around tokenomics, around productivity, and and business impact. Then we think about, you know, the the the way in which this surfaces within your organizations. Again, you want your teams to be able to run faster continuously, and AI is certainly proving that it can enable that. And being able to effectively activate the the the key differentiator in your business that is your unstructured data allows them to run at that speed. AI is getting smarter by virtue of the cycle that it goes through. So we talk about this idea of compounding productivity. As your teams work faster and they're leveraging AI to deliver better results, those get captured within the underlying data that gets fed back into AI, and that's where you get this compounding effect. At the end of the day, that's how you start to drive these these, you know, 10 plus x, returns on investment for the business. So last slide here. And I think we're, hopefully running a poll. I I... If I... I think I saw it go out. So please make sure that you do, respond to that poll. It can be really useful in terms of understanding kinda how we can best follow-up and provide additional detail to you. But this is the last slide I'll talk about because the key piece here though is, okay, we're an underlying platform, but we fit into an ecosystem of solutions that you leverage today. And so this is really important. Within my organization, we look very, very closely at integration with third party solutions, whether that's as you think about our ransomware protection service and being able to feed into your Siemens source solutions like CrowdStrike, like Splunk, like Microsoft Sentinel, for example, so that we can provide a better together scenario there. You think about DriveIQ, as I talked about, is another great example where it can integrate into all those other unstructured data repositories and content repositories that live around your data on the SUNY and being able to provide not just the understanding and visibility, but the governance and cost savings on top of that. So, you know, you'll continue to see us here building out examples of solutions and how customers can unlock the value of what SUNY provides in the environments where you... Where you're living and working. Again, whether that's in the hyperscalers, whether that's in homegrown solutions that you're building on prem. Alright. So with that, I think, we're gonna use the last few minutes here. I've seen some questions come in over q and a, and so I'll I'll go ahead and tackle those. So the first one I see is my board is asking about AI risk and returns. What can I take back that shows other companies are doing this? This is a great question. And this is something we talk a lot about with our members of our customer advisory boards at our Nasuni verse events. And what I'd say is one of the things that we see across the board, and frankly, it was one of the the big reasons why we went down the path of looking at DriveIQ and the important role it could play, is that governance has become, you know, just front and center in all of these conversations. And we see that surface in a variety of ways. Some customers are fairly mature and that they've defined an AI council under various different names. And, frankly, that is populated by c suite members, and they're talking about how do we define what AI readiness means and what governance means, and then looking to actually have a way to implement and automate that in a platform like DriveIQ. For others, that might be more distributed within different organizations, but nonetheless, still making sure that there are auditable, guardrails and checks and policies in place to ensure that the data that is the gold mine that your your business runs on isn't exposed, unknowingly, and in particular, that you have that control in automated fashion, because the reality is AI is definitely still spread across different environments, and I think everyone knows about, you know, bring your own AI is definitely still a reality. And so you've gotta have a platform and a framework that allows you to protect against that risk. Great question. So the next one. Four new products, which do I start with and in what order? Ah, this is a great one. So if we kinda come back to the first slide that I I talked about and you saw those four phases, understand, manage, protect, and activate. You may be in different in different, phases of that of that progression. Many of our customers, that live... Work with us today, they're all the way out to that point of thinking about activation of that data. But we really can meet you where you are. You may have a solution that provides, you know, content level visibility today, and you're really focused on, alright. Now I wanna be able to unify that in the single global namespace. Well, great. The core is a foundational place. You may be earlier in that, and you're just trying to wrap your arms around what data do I have. A lot of our prospects and customers are, grow through mergers and acquisitions, and that's a great use case of, hey. We just acquired this company. We have no idea what the data is. We don't know who has access to what. That's where DriveIQ really can be a great starting entry point. You're not even having to think about migration of that data to a newer platform yet. You're really just focusing on understanding that data, optimizing that data both from a security and risk perspective, but also from a cost perspective. And then you can build a strategic plan for what data should move over to a platform like Nasuni, go direct to archive, etcetera. Another great question. Alright. Next one that I see. We already have Copilot in our Microsoft stack. Why would we add Nasuni to get a access to our files? Ah, this is a great one. So, yeah, you're exactly right. A lot of our customers are big SharePoint users, m three sixty five users, and you can get a great experience, particularly with some of the new capabilities that Microsoft has rolled out around Copilot leveraged against that SharePoint data. But the reality is, you know, our customers' SharePoint is one, data repository, but it's all of the other data that does not belong in SharePoint. All of those critical design files, that geospatial data that I was talking about before. Even though during COVID, share... Share form became a little bit of a dumping ground, the reality is you, you know, you wanna be able to leverage this across all of your unstructured data. And so that's where we really see it as complementary. And so if you think about leveraging Copilot or you're building out an agentic workflow, AI activate, as an example, can plug right into that workflow. And now you've got, you know, an agentic solution or an LLM that has access to all of the knowledge and and and insight that's available across those two different, data repositories, but you're not having to compromise and pick one for the other. So the level of scale at which we work, the types of data that we that we manage, protect, and activate, that's really all of that data that lives outside of that SharePoint ecosystem typically. But one of the other key things is coming back to that kind of where do I start. A lot of our customers, you know, for them, SharePoint did become a little bit of a dumping ground, and they struggle with overages for SharePoint, which can get very expensive. And so one of the great things about DriveIQ is it can provide a phenomenal discovery and and archive solution where it can identify based on whatever, terms you define in terms of what is inactive data. Maybe that's, you know, all SharePoint files that haven't been touched in two years. It can archive those, leaving behind a placeholder. So... And users are always able to find their their... The files that they would expect to if they did wanna come back to it. But now the underlying files themselves are on Nasuni. They're protected there. And really important, they're still available to AI, as well through something like AI activate. Alright. So there's a bunch of other questions that, unfortunately, we won't be able to get access to, but I do, wanna say that we'll follow-up. We'll make sure to to, connect with folks, on the questions that we didn't we didn't... Or didn't have time to answer. And let me just wrap up by saying, thanks to Dave for joining me here today, and thanks to all of you for spending time with us. We know how, precious your time is. Really appreciate it. Hopefully, this was useful to give you a little bit more insight in terms of what we're up to at Nasuni, and I'll hope to, run into one or all of you, soon after this event. So thanks very much.