AI Agents Are Failing and It's Almost Never the Model's Fault | Alberto Pan, Denodo

After two years of AI pilots, enterprises are finally diagnosing what went wrong, and the answer keeps coming back to data. Alberto Pan, CTO of Denodo, joins Craig Smith to walk through the findings of the company's AI Trust Gap Report: a survey of 850 enterprise data leaders that reveals the dominant failure modes of enterprise AI agents are almost never the model's fault. They're caused by stale data, missing context, and inconsistent semantics across the hundreds of data sources agents need to access to do real work. Pan explains why traditional data warehouse and lake house architectures - built for analytics, not real-time decision-making - are creating an invisible ceiling on AI performance, and how Denodo's logical data management approach lets agents query data where it lives without centralizing it first, while enforcing consistent governance across every source in one place. The conversation also identifies two specific traps most organizations fall into as they try to scale AI - over-centralizing data into a single system, or building custom ad hoc data layers for every agent - and why both approaches collapse in a multi-agent world where agents need to cooperate, share context, and work from a common semantic foundation. Subscribe to Eye on A. I. for weekly conversations with the people building and deploying the future of AI.

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:00Craig S. Smith?

  The adoption by enterprise of both generative AI and agentic AI has been much slower than a lot of people anticipated. They don't trust the AI.

• 0:12Speaker 01

  Most companies, most organizations, as you know, during the last two years, they have been doing pilots around AI, right?

• 0:18Craig S. Smith?

  How long is it going to be before all large enterprises are using agentic AI in all relevant processes?

• 0:27Speaker 01

  I think it will take significant time because at the end of the day, there are bottlenecks that in many cases are related to organizations or even legal reasons in some cases.

• 0:35Craig S. Smith?

  Can you talk about the trust problem, what you guys, I think, call the trust gap?

• 0:44Craig S. Smith?

  Why don't you start by introducing yourself, how you came to Denodo, what Denodo does?

• 0:51Speaker 01

  Yeah, absolutely. Well, first of all, thank you and thank you for having me, Greg. So yeah, I am Alberto Pang, I am Chief Technology Officer of Denodo and also a member of the founding team. And Denodo is a global company today. We are headquartered in Palo Alto and we have offices in more than 25 countries. But our story actually began in A Coruña, which is a small city in the northwest of Spain. And actually, I am still based there today. I am talking from A Coruña today. And here, basically, I am leading our R and D team. The majority of our R and D team is based here.

• 1:33Speaker 01

  And about my journey, my journey actually started in academia. I was for many years, I was for many years doing research on data management. And actually, the core technology that powers Denodo today grew out from that original research that me and other members of the founding team of Denodo were doing.

• 1:55Craig S. Smith?

  And Denodo, describe the problem that you guys set out to solve and then the solution you've come up with.

• 2:03Speaker 01

  Denodo, first of all, is a data management company that enables organizations, typically big organizations, to create a unified, real-time

• 2:15Speaker 01

  Access layer, data access layer across all their data sources, right? We typically call this a universal semantic layer because it provides the data in the language of the business. So it makes it easy for people and of course also today for AI agents, not only to get access to the data that they need, but also to understand how this data should be used in different business contexts, right? And a big difference between the NODO and traditional data management architectures is that the NODO does not force you to consolidate everything upfront

• 2:57Speaker 01

  in a central system, like in traditional data warehouse or warehouse architectures. With the NODO you can query the data where it lives. And this has a number of benefits. First, it removes the bottlenecks that are typically associated to centralization, also allows accessing the data in real time where it lives, as I mentioned, and also allows users and agents to get access to all the data.

• 3:24Speaker 01

  Because typically in many organizations, what you have in the central systems and data warehouses or warehouses is only a small percentage of the data. So with the NODO you can actually get access to all the data. And we call this typically a logical data management because, well, you don't need total physical data replication, right? And the underlying technology that allows this is called in the market typically data virtualization.

• 3:57Craig S. Smith?

  And the semantic layer that allows both humans and AI agents to make queries in natural language, or how does that work?

• 4:11Speaker 01

  Well,

• 4:13Speaker 01

  Natural language is one of the of the interfaces supported and obviously we are using GenAI for that, but actually you can also consume the data with more traditional interfaces. For instance, actually, what we try to do is to deliver the data so it can be consumed with any tool. So for instance, if you are accessing the data products that you create with the NODO, with tools like Power BI or Tableau, you you will probably access them using technologies like JDBC or ODBC, right? Technologies oriented to SQL. But if you are a data scientist and you want to access the NODO data with a notebook, maybe you will use a technology like Arrow.

• 4:54Speaker 01

  Or if you are creating a web application on top of the data products exposed by the NODO, maybe you will use REST APIs, right? Or if you are an agent, you will probably use MCP. All those are examples of interfaces that are available and that you can use to to query these data products that you create with the NODO across your data sources.

• 5:19Craig S. Smith?

  But the semantic layer in order for the interface to know what data you're trying to access in disparate locations, does that unify not only access but the language that you need to access? I mean, I've spoken to a lot of people about data ontologies and maybe one data store calls a customer, a customer, and another data store calls a client, you know, something like that. And you need to link the two. You mentioned graph databases, is that what you're saying?

• 6:01Speaker 01

  That problem of solving inconsistencies across data sources is very, very important in practice and is crucial in the NODO. And yes, the NODO provides ways to explicitly specify the meaning of each term and each piece of data in each data source and also even to consolidate.

• 6:23Speaker 01

  Different formats in a unified format, right?

• 6:28Craig S. Smith?

  Right.

• 6:28Speaker 01

  So you can go either way. Either you can either expose in a consolidated data product, the data in a single format. Or you can also say, no, I have these different data products. And the way, for instance, that this KPI is computed in this data product is this one. And the way it is computed in this other data product is this one, right? So the agents or the users accessing those data products have the full context to understand. Okay, for instance, what data product should I use in this particular business context?

• 6:59Craig S. Smith?

  You describe it as a semantic layer. Is that software that sits in the cloud? Is it software that sits on premise at the organization?

• 7:17Speaker 01

  Well, actually, both options are available. I would say that today, most of our customers use the node in the cloud. But actually, the node can be deployed at any location, on premises, private cloud. Public cloud, if you're running the node in the cloud, you can also choose between a SaaS model, where basically we manage the infrastructure for you, or you can decide to manage the infrastructure yourself. So we provide a lot of deployment options. We typically work with very big organizations. So I think for this type of organizations, giving them a flexibility in the type of deployment is very important, right? Because, for instance, sometimes they are accessing very sensitive data.

• 8:02Speaker 01

  So maybe they want that for this particular use case, they want the node to sit on premises data, because that data is not in the cloud. So we really try to give that flexibility to our customers.

• 8:14Craig S. Smith?

  How then has evolved with Agentic AI?

• 8:19Speaker 01

  Probably the main change between Agentic AI and the initial iterations of the Agentic AI applications is that, well, we have just mentioned a minute ago, right? If you think about it, the first thing A applications were for data analytics. So they were about helping humans to get information, to make better decisions.

• 8:47Speaker 01

  And now we are seeing a new phase that maybe we could call them agentic AI or operational AI, where basically agents are being embedded directly into the into business workflows. I don't know to manage, I don't know, customer incidents or insurance claims or logistics. So this is just, it's not longer only analyzing data. It's real time decision making. And I think that's where traditional data management architectures start to break down. Because if you think about it, traditional data management architectures, data warehouse, the data warehouse and so on, they were built for analytics.

• 9:29Speaker 01

  So while AI remain confined to analytics, I think this was mostly fine. Obviously, there are also limitations there, but it was mostly fine. But when we are starting to extend in the scope, I think that's that changes things significantly. And that is having a big impact also in the, I think, in the node and in the perception of the node by the market, yeah.

• 9:53Craig S. Smith?

  Adoption by enterprise of both generative AI and agentic AI has been much slower than a lot of people anticipated. And it's primarily a trust problem. Because enterprises are reluctant to integrate generative AI or agentic AI into their systems beyond maybe a customer facing chatbot. But even that is can be problematic. They don't trust the AI. And this is particularly a problem with generative AI, which is probabilistic.

• 10:38Craig S. Smith?

  And, you know, older. Systems certainly from the expert system era of AI were deterministic and a lot of enterprises feel much more comfortable knowing what the answer is going to be, or the answer is going to be deterministic. Can you talk about the trust problem, what you guys I think call the trust gap? And then we'll talk about the study that you just came out.

• 11:11Speaker 01

  This study basically was a survey on more than eight hundred, more than eight hundred and fifty

• 11:20Speaker 01

  data leaders in big organizations with representation of companies from the US, from EMEA, from APAC, and all main sectors, including financial services, the public sector, healthcare, you name it.

• 11:39Speaker 01

  Focus specifically on big organizations, organizations with more than 1,000 employees at least. And yeah, what we found is consistent across industries and also consistent with other industry surveys from the likes of McKinsey or Garner. And it's that most failure modes of AI agents today are related to data. Most companies, most organizations, as you know, during the last two years, they have been doing pilots around AI, right? And now, I would say in the last few months, they have been analyzing the results. And I think most of them have reached this surprising, surprising conclusion

• 12:21Speaker 01

  that is typically, it's not that the model is not smart enough. It is that it does not have access to the right data, or maybe the data is not up to date, or the agent is not understanding how the data should be used in a particular business context, right? And that, that, that is what creates hallucinations. That is what it creates when the agents overreach what they should do. And this is basically what is creating this, this trust problem, right? We drill down in the, in the report. There are some numbers that I would say, and we try to understand exactly what are the root causes of the problems. We get some striking numbers.

• 13:03Speaker 01

  For instance, almost seventy percent say that the lack of real-time data for the agents has been a problem. And the percentage is even bigger if you take organizations that actually have AI in production. So that suggests that, you know, as organizations go into production, they realize that this is a real problem. Also, another problem that is mentioned in the study is that in these organizations, AI agents need data coming from an average of four hundred data sources, and more than eighty-five percent of organizations require more than one hundred data sources. So these

• 13:46Speaker 01

  numbers are amazing. And almost seventy percent also report problems, probably because of this data distribution, report problems getting agents to comply with consistent governance policies and consistent and consistent work rates, right? So I think that's the main conclusion that you can get from the report, that the reason why there is this trust gap, the reason why the agents sometimes overreach or providing consistent results are mostly data problems.

• 14:23Craig S. Smith?

  Can you break some of those down? Live data, for example, is that data architecture problem?

• 14:33Speaker 01

  I would say it's an architecture problem, because as we were mentioning before, initially, AI was focused on analytics, so it was based in these data warehouses, data lake house systems that force you to first copy all the data there. So when you have a data replication process, you always have some latency. So the data will never be one hundred percent real time. That's fine for analytics, typically, when you are trying to find trends or things like that. That, eh, but it's not good enough for many eh eh business workflows, right?

• 15:10Speaker 01

  So that is why, let's say, the old foundation for AI is is is not good enough to to get eh real-time data. As I said before, seventy percent of the of the organization said that even data that is one minute stale, you know, is is is not valid for some AI agents.

• 15:32Craig S. Smith?

  How does the the Denodo semantic layer um correct that or or enhance that?

• 15:41Speaker 01

  Yeah, because with with the with Denodo, you don't need eh to centralize the data upfront. You don't need to replicate the data. Denodo is able to query the data where it lives, right? So you can think about this data product that you create with Denodo as virtual data product. So maybe one data product is created with your Laehouse and is pointing to the Laehouse. But maybe this other data product is pointing to your CRM or an operational database, right? Where data is updated live. So eh actually, the data product that you create with Denodo can provide this data up to date in real time. Well, for instance, if you are using a centralized architecture,

• 16:23Speaker 01

  you first need to copy the data from that operational database to the central system. And that will always introduce a latency, so data will not be in real time.

• 16:32Craig S. Smith?

  The right data uh is obvious uh that you need the right data for for the right uh action or answer. But what what's the issue there? What creates the trust gap in traditional architectures?

• 16:50Speaker 01

  There are several issues here, right? Because having the right data means different things. First, first, you need access to all the data. And if you are confining your AI applications to a single system, for instance, to a data warehouse or a Laehouse, then you will probably be missing part of the data that is needed. So that is one part of the problem. And the second part of the problem is this semantics that we were talking about, right? First, that the the semantics need to be described, and second, the semantics need to be consistent. To explain the problem of semantics, I sometimes use an example, right? Imagine that you have a very brilliant DBA, a new hire,

• 17:35Speaker 01

  with perfect SQL knowledge, but it's a new hire, so lacks the business context, right? If you ask them for, I don't know, for instance, in the healthcare space, the readmission rate of cardiac patients, and then you give them access to the raw tables that contain the data to answer that,

• 17:57Speaker 01

  but you only give them that, they will fail. They will fail because they won't know, for instance, which specific diagnostic codes define all the different cardiac diseases, or they will not know what is the exact formula for calculating a readmission in this case. For instance, if a patient is readmitted for a different diagnostic, is that a readmission or not, right?

• 18:21Speaker 01

  They might even have problems knowing what data is in each column table, right? So without a semantic layer to define these business rules, and also in a consistent way across all data sources, even the smartest expert is forced to guess. And in the world of AI, guesses are basically hallucinations, right?

• 18:44Speaker 01

  So that's what we mean with the right data. First, having access to all the data. Second, the data with the right level of freshness. If you need this piece of data in real time, it should be in real time. And third, with consistent semantics.

• 18:59Craig S. Smith?

  From the user's point of view, is it a dashboard? Or is this all happening unseen to the user?

• 19:08Speaker 01

  The NODO is more like this catalog of data products.

• 19:14Speaker 01

  It's like a web commerce, you know, like Amazon or a digital storefront, where the node doesn't actually move the data to create these data products. It's like a catalog of these data products. These data products maybe come from a cloud database or on-premise system or a SaaS application, right? And then these data products can be consumed, basically using any interface that you want, right? We actually also have a data, our own, let's say, data marketplace, which is a web application that human users can use. But also, as I mentioned before, these data products can be accessed with any tool, any tool that want to consume that data using all those interfaces that I mentioned before.

• 19:58Craig S. Smith?

  On these foundations of trust with the AI, you know, we talked about live data, the right data, and then guardrails. Can you talk about the guardrails that are necessary, why they're necessary?

• 20:16Speaker 01

  Yes, yes, absolutely. Well, I think, you know, there are some types of worries that everybody understands. Everybody talks about, I don't know, things like ensuring that the prompts are safe,

• 20:29Speaker 01

  protecting yourself against prompt injection, forbidding the agents the access to, I don't know, to certain tools. Everybody understand that, right? But I think what sometimes goes under the radar is that, as I mentioned before, and the study shows, agents need to pull data from a massive variety of data sources, right?

• 20:54Speaker 01

  And when you have to work with multiple data sources, then you have several problems, right? For instance, the problems of the consistent semantics that we mentioned before. But also, how do you enforce consistent security policies across all those data sources? So, for instance, imagine that you have a customer support agent, but because of GDPR or any other regulation or whatever, the agent is not allowed to see certain data from customers, right? And that data from customers will not be, in many cases, may be available in several data sources, in several places, right?

• 21:40Speaker 01

  So if you don't have something like the NODO and the agent needs to access those different data sources, you will need to implement those security rules in several places, right? With the NODO, you can implement that ruling only one place and you can be sure that it will be enforced for all the agents, for all the data sources and so on, right? Also, I would like to briefly mention or briefly comment why agents need to access to so many data sources, right? Well, for many reasons, right? First, because organizations are very complex, data is distributed in many places. Also because real-time data sometimes prevent consolidation, as we mentioned, because you need to query the data where it lives.

• 22:25Speaker 01

  But a third reason, and this is very fundamental about agents, is that agents are unpredictable by definition. So by definition, agents, we use agents when we don't have a predefined set of rules. Where we have a task, when we have a task where a predefined set of rules does not cover all the scenarios, right? Because if we had a predefined set of rules that covers all scenarios, we could use a conventional workflow. So the whole point of an agent is that it needs to find solutions for problems that they have never seen before in exactly the same form.

• 23:05Speaker 01

  So and if you think about it, this means that we don't know in advance exactly what data an agent will need. In a typical workflow, we know, okay, this workflow needs this piece of data, this piece of data, and we can expose only that piece of data. But with an agent, you have to give them access to all the data that potentially may be relevant for the task at hand. And that also contributes to increase a lot the number of data sources.

• 23:34Craig S. Smith?

  You were talking about live data. Why is live data becoming critical for trust for the AI? What breaks when AI acts on stale data?

• 23:44Speaker 01

  Basically, that leads to bad decisions or has the potential to make the, to cause the agents to make bad decisions. For instance, if your agent is talking with a customer about a service down incident. It needs to know the state of the service now, not five minutes ago. Or if it is deciding how to locate a certain resource,

• 24:13Speaker 01

  for instance, a seat in a plane for an upgrade or something like that, right? It needs information about the resource availability now. The resource availability 10 minutes ago, one hour ago, may not be good enough. Not all workflows are like this, right? There will be agents that maybe will not need real-time data. For instance, agents that work only for analytics, as I mentioned before, sometimes don't need real-time data or don't need... But there are many, many workflows that actually need real-time data. In other case, the agents will not be able to make good decisions.

• 24:48Craig S. Smith?

  The challenge of enforcing these guardrails and governance across distributed environments,

• 24:59Craig S. Smith?

  how do you handle that? And how does Agentic AI make that even harder?

• 25:06Speaker 01

  Since the NODO is this single entry point for agents to get the data, or these data products are the single entry point to get the data, you can be sure that those policies will be always enforced, right? Behind the scenes, maybe the data for that product is coming from several data sources. But the agent never interacts directly with the data sources. It always goes through this virtual data product offered by the NODO. And it is there where the security policies and the guardrails are enforced, are defined first and then enforced.

• 25:42Craig S. Smith?

  And these problems that we're talking about, underlying the trust gap that you guys see

• 25:51Craig S. Smith?

  in enterprises regarding integration of AI?

• 25:56Speaker 01

  You know, the organizations were through several stages in the last two years, right? The first stage was, okay, this is very cool. Let's start to experiment, let's start running pilots.

• 26:07Speaker 01

  Then the second stage was, you know, wow, this is, this is, this sometimes is, you know, this makes for impressive demos and sometimes actually

• 26:19Speaker 01

  provides fantastic insights or fantastic suggestions, but also very often makes these strange decisions or, or, or overreaches, maybe the agents overreaches, those things that it should not do. And then, wow, and initially, as you mentioned before, this is not an expert system, this is not symbolic, neuro-symbolic. So at first, I cannot know what went wrong, right? So I think that that was the second stage. The third stage was, okay, we, we are investing a lot of money on this, let's analyze what's going on. Let's analyze what's going on. So first stage is, okay, let's invest in auditability, let's invest in traceability and auditability.

• 27:00Speaker 01

  Because actually, those agents are, you know, using external tools and so on, you are quite able to know exactly, okay, I got this data, then I analyzed the data, in function of this, I decided this and so on. So actually, you can get good traceability if you really invest in it. And then analyzing these failure modes, it's when I think these organizations discover these gaps, that are mostly data gaps, right? And I think that changes the strategy. For instance,

• 27:33Speaker 01

  I, I, I was in, last week, a very recent report from IDC, from the, from the, from the, and they were, I was actually comparing the report from 2026 and the report from 2025. The report from 2026 was released, I think, in April, and the other one, I think, is February 2025, right? And the question was, what is the main barrier, or no, what is your main priority for investment in AI? And in 2025,

• 28:01Speaker 01

  revamping your data architecture for AI was like in fifth place or something like that, or sixth place. And now it's the first place, right? And I, I think that changes because of this realization that, that where the, the failure modes really, really are.

• 28:19Craig S. Smith?

  That's interesting. So we're at a point

• 28:23Craig S. Smith?

  in, in the adoption by enterprise where there was the pilot phase, then they took it.

• 28:34Craig S. Smith?

  A little further and ran into these problems, understood that it's data. And that's where Denodo comes in to unify the disparate data sources and then provide these unified, universal policies. How should executives think about balancing existing lake house investments with the new layers that Agentic AI requires? And what organizations, what should organizations do as they scale AI?

• 29:17Speaker 01

  Yeah, well, I think that for most organizations, the lake house is part of the solution, right? It contains a lot of valuable data, is a great foundation for data analytics and data science. But I also think it's not enough. Even in the world before AI, there will always be data that is not in the lake house. You will sometimes need to access real time data. You will need strong semantics across several data sources. So even in the world before of AI, we think that lake houses were not the final answer. Very useful, but not the final answer. But with AI, all those problems are multiplied.

• 30:01Speaker 01

  And that is why we think that this new data foundation is needed. So our recommendation would be leverage the investments that you have, the data sources that you have, but prepare this common infrastructure, right? Data infrastructure that goes beyond central system, that actually covers all your systems and ensure that your semantics, your governance and so on is defined there. And this is also very related to.

• 30:33Speaker 01

  What organizations should avoid, what organizations should not do? Well, I see a lot of companies falling into what I call the ad hoc trap. Meaning that, you know, since, because these traditional architectures have limits for the new use cases. And there's so much pressure to deliver results. It's tempting to take shortcuts, right? It's tempting, for instance, to create an ad hoc data layer just for one specific AI app. And then another data, a specific data layer only for this other AI app, right? Or you may say, okay, this agent will be a customer support agent.

• 31:18Speaker 01

  So I will restrict it to work only inside the CRM data, right?

• 31:24Speaker 01

  And that might give you a quick win, but we really think it's a dead end. Because we are rapidly moving towards a multi-agent world. And in this world, agents will not work in isolation. They will need to collaborate. And if every agent is built on a different data silo with its own definitions. They won't be able to talk to each other, they won't be able to cooperate. That's why, you know, those are the two traps that I see for organizations. The first thing is, okay, let's try to create AI on a single system. We will try to centralize all data here. That will never work for big organizations.

• 32:05Speaker 01

  And the second trap is, okay, since the first thing is not working, let's do ad hoc solutions for every agent. And that will not work in the long term either, because agents need to cooperate and talk to each other.

• 32:16Craig S. Smith?

  And these other solutions, centralizing the data or building data layers for each AI or agentic application, is, are companies, enterprises, before they find a node, are they doing that themselves? Are they using consultants? Is that a solution that they come up with? Because it's, it seems an obvious fix? Or are there people in the market telling them to do that, that that's the solution?

• 32:53Speaker 01

  A bit of everything, right? Obviously, on one hand, obviously, you know, it's vendor tries to.

• 33:00Speaker 01

  Bring the data to their systems and to have their AI engines work on data, right? So obviously, the vendors try to do that. But if you think about it, where does that end? If all of them are partially successful, then at the end, what you have is many silos, right, with different AIs that cannot talk to each other. So part of it is the market, of course, but part is also a natural evolution of things, right? So you start, for instance, as I said before, the first AI applications that you do are in analytics. So you say, it's natural to think, okay, I will use my analytic systems for that or my main analytic system for that.

• 33:42Speaker 01

  And then only later you realize, oh, this is not enough, because now my agents go beyond analytics. And then when you realize that, you say, wow, I have a lot of pressure to deliver, right? Because I really need to deliver this. And then the solution is, okay, I will do something ad hoc, it will work for this, and let's see what happens, right? So there are also very, very natural, very natural trends. But I think that if you think long term, you see that, you know, you need to go beyond that.

• 34:17Craig S. Smith?

  This is happening quickly. This, the trust gap has formed quickly and then, you know, companies are building these systems quickly and the capability of the systems is advancing quickly. How does Denodo keep up with all of that? Is your product also evolving?

• 34:44Speaker 01

  Well, I think this is something that obviously is not happening only to Denodo, also to... To everyone, right, in the space. If I, I am, as I, I am the technology officer, right? So I own the roadmap of the product. And if I see my roadmap, I don't know, five years ago and now, wow, now it's completely dominated for AI, right? In those two flavors that I mentioned before, what AI can do for the NODO, what the NODO can do for AI, right? And to, to be honest, this is because also we see the trend as unstoppable. And we are seeing, for instance, the evolution of our users and our users.

• 35:23Speaker 01

  And five years ago, probably this would have been unthinkable, but we are seeing like, okay, now maybe we are starting, you know, the, the number of non-human users is growing so fast, right? That actually that changes a lot of things, that changes a lot of things about your product and about how you should expose the data, how you should represent the data. And it's of course a learning process because as you said, things are, are changing so quickly that we are all learning, right? Of course, we are all learning. And as I said before, I think these lessons that we were discussing were the lessons that many organizations got after this one year or two years of pilots.

• 36:03Speaker 01

  And now I think we are entering a stage of more maturity, but there will still be very significant changes because the landscape keeps, keeps moving. Yeah, in, in our website, if you go to, to denodo.com, you can download the report from there. Yeah, and also learn more about, about Denodo.

• 36:21Craig S. Smith?

  From Denodo's point of view, has this, the, the findings in this report, has it changed the way, you, you, you talked about the, the roadmap, map that you oversee? Has it informed your future? You know, when you define the, the trust problem?

• 36:48Speaker 01

  Yes. For instance, some of the topics, for instance, live data access probably was even more important than we were anticipating, right? So, for instance, that has had an immediate impact, right, on optimizing even more those type of accesses, right? So that's only one example.

• 37:13Speaker 01

  I think we were already pretty strong on those particular areas, right? But for instance, that would be an example where we were actually were surprised about how the strong, how strong the results were. And that has, obviously, also has an impact on what we do.

• 37:28Craig S. Smith?

  Are you guys, is this a period of growth for you guys now that everyone's adopting AI?

• 37:38Speaker 01

  Yes, well, the Noto, you know, is a consolidated player in this place. We have been in the data space for many years. And for instance, we have been leaders in the data integration, Magic Quadrant from Garner for six or seven years in a row, something like that, right? But certainly, AI is being an accelerator for us, is being a clear accelerator for us. I think it's because those changes that we have been discussing, right? And that actually, we think that fit very well with our original vision. So yet, especially in the last year, AI has been a great accelerator for us, yeah.

• 38:18Craig S. Smith?

  What about a sort of general adoption? You've identified the data problems that create the trust gap that have been slowing adoption. If that's solved with the Noto, do you see AI adoption? People ask me all this all the time. In how long is it going to be before all large enterprises are using Agentic AI? And in all relevant processes and?

• 39:00Craig S. Smith?

  Uh, yeah. Is it five years, ten years, or, or, uh, or less?

• 39:06Speaker 01

  If we say all organizations in all relevant processes, I, I think it will take significant time because at the end of the day, there are bottlenecks that in many cases are related to organization or even legal reasons in some cases, right? But i- if we lower the bar, the bar a little bit and, and we say, okay, let's say the twenty-five percent most advanced companies using it in critical workflows, critical workflow, let's say twenty percent of this, their critical production workflows, I would say by the end of twenty twenty-seven, and, and maybe I am being, well, you know, oh, you know, eh, I, I was going to say that maybe I, I, I am being too conservative

• 39:51Speaker 01

  because I, I see the, the thing, the, the things changing, changing quite, eh, fast. But it's also true that at the end of the day, in big organizations, everything takes more than expected. Yeah, because there are, you know, all types of bottlenecks, not only technology bottlenecks. From the technology point of view, to be honest, I think the technology is ready, the models are good enough for most of the use cases that we are considering. I think with solutions like this, you can also solve the data problem. So I think that from the technology point of view, the technology is mostly there. Eh, also, data ability has improved a lot.

• 40:31Speaker 01

  I think the technology is mostly there, so that's why I think at least the most advanced companies that are most advanced, more mature with technology, it will go quite fast. Actually, a- a- another thing that you see in this survey, and also in other surveys, like the, the IDC survey that I mentioned before, is that actually the companies with a high level of data maturity, for instance, that already have a well-established eh, data product foundation and so on, actually, actually in those ones, the level of adoption of AI in production is already significantly higher. Right? So eh, I think technology probably is not the bottleneck anymore.

• 41:14Craig S. Smith?

  Yeah, yeah, that's fascinating. Um, okay, is there anything that I haven't?

• 41:22Craig S. Smith?

  Touched on that you'd like listeners to hear?

• 41:26Speaker 01

  Um, no, I think, you know, it was quite comprehensive, right? Yeah. Hopefully it has been also useful for the listeners.

• 41:34Craig S. Smith?

  Okay.