Techstrong TV March 10, 2026

Hey, everyone. Welcome back here to Techstrong tv. Let me introduce you to Dr.

IA Chenko. Uh, Dr. Chenko is the CEO of Immuno Web.

That's I-M-M-U-N-I web, WEB, Dr. Chenko ia. Welcome to Techstrong tv.

It's great to have you on here. Thank you very much, and thank you for having me here today. Absolutely.

Um, let's start off with little bit ili, if you, I'm gonna call you ili, if it's okay. Ilea. Let's start off with a little bit about you, right?

Your PhD, CEO of immuno web thought leader in, in the space. Tell us a little bit about, about your story, your life story, if you don't mind. Certainly.

So, probably my, uh, story is not very usual because I'm a cybersecurity professional. I've been in cybersecurity practice for about 20 years. That's another way to say that.

I'm an old man, uh, And I got your beat, so go easy, Uh, right. And, uh, actually, when, um, I was in my, uh, thirties, I, uh, decided, uh, to start studying law and I became a lawyer. Uh, and, uh, basically speaking today, uh, I'm an attorney.

Uh, my practice is mostly focused on complex, uh, cyber law cases or cases that involve, um, cyber crime investigations. And I also have a pleasure and privilege to only lead the amazing technical teams at Muni Web. Uh, in a nutshell, what we do, we're a global application security company, headquartered in Gene Switzerland.

We also have regional offices in London, Washington, DC and Dubai. And, uh, we provide, uh, cybersecurity services, uh, to more than 1000 enterprise customers from 50 plus countries. Really?

That's interesting. Um, so I did the opposite. I went to law first, hated it, and then got into technology, and that was about Jesus, 35, maybe years ago, maybe more.

And, um, never looked back though. You know, what you learned in law, you always stays with you, right? It's the way of looking at things, a way of thinking about things, way of approaching problems.

Um, but it, you know, cybersecurity has become one of those areas where having a law degree is a good thing, right? Because the, the line between legal illegal, wrong, you know, morally wrong versus legally wrong, is, is often there's a wider gulf than one would think, right? Um, so it, it's an interesting thing.

So you didn't found Muni web, or you did, I did You found an Muni web to kind of be the vehicle around this, this whole law thing? Or, or did the law thing come after? I would say probably it happened simultaneously, but at the media, web, we're certainly not a law firm.

So we don't, uh, provide legal services, right? However, we have a lot of different products where we help our customers to better understand, uh, for example, their regulatory duties relating to cybersecurity, data protection, privacy, incident response. Now, we have a lot of different, uh, products relating to security and safety of artificial intelligence products and technologists.

Uh, so basically speaking, we are pure tech firm, but we try to provide some additional insights to our customers. We're not competing with lawyers, not at all, but we try to bring in a certain areas that, uh, usually stay in darkness and nobody cares till they have a lawsuit. We try to bring them to our customers, say, listen, probably you, you, you should look at this specific item because it's might trigger an issue, and probably talk to your general counsel.

So we try to put the cherry on the cake. Got it. Excellent.

You know, you mentioned artificial intelligence, ai, ai, I mean, these last couple weeks have just been historic in what we're seeing at the AI starting to do now, probably application security. AppSec has been one of the most, uh, impacted domains, uh, from ai, you know, the release of Claude Opus and finding all these vulnerabilities and open source projects, the idea that not only is AI helping you generate the code, but it's actually testing code. It's testing code as good as some of the best, you know, SAS or, or, or, you know, I don't wanna say das yet, but certainly static tests out there, and a lot of people are, you know, foretelling and, and contemplating, and pontificating maybe is a better word, on what the future of application security is in, in an AI world here, right?

Where not only does it write the code, but test the code and potentially fixes the code, right? And, and that's a lot of, it's a lot of responsibility to put in front of what's still a pretty unproven, I think, uh, technology. I'm wondering from where you sit, where do you see that Dr.

Chenko? So, I believe that, uh, my position, uh, and my view on the future, uh, of artificial intelligence, um, uh, have been consistent since 2019. Uh, this is actually when, uh, at the web we received our first, uh, international award for practical use of machine learning and ai.

So it was, uh, before the launch of chat, GPT, and basically speaking, my opinion is still the same. So AI is an amazing tool for lawyers, for engineers, for penetration testers, for many other people. That is an amazing tool.

But that's it. I don't think that, uh, artificial intelligence will ever be able to fully replace human experts. Uh, I acknowledge that in many organizations, we had too many people doing redundant tasks for some good or bad reasons.

And currently we also see what we call AI washing, when a lot of people are being fired, uh, uh, under the pretext of, uh, AI automation. But in reality, organizations truly never needed this kind of roles internally. And, uh, with this AI trend, they start rethinking, reviewing, and actually deciding, saying, well, if we can do this task, uh, with a team of two, why actually we had 25 people, seven supervisors, uh, three additional supervisors, and the entire team of technical support, we just don't need it.

And, um, talking about practical implications, uh, of, uh, artificial intelligence. Some people say that AI will replace virtually everybody, alright, but the, but the challenge is that if you don't understand an output of AI tool or solution being at the complicated m and day contract or example, or source code of a sufficiently complicated, sufficiently sophisticated, uh, software, if you, as a lawyer or non-lawyer, or as an engineer or non IT person, you don't understand that you'll probably have a contract that will do exactly the opposite of what you initially contemplated, uh, to do. Same with the software.

If you just copy paste, uh, output from chat GPT or Claude, you'll probably create unreliable software to put it mild. Uh, so we still need, uh, human experts to understand, to utilize ai because we cannot just say, listen, I want to create a Microsoft office. Uh, we will not have it.

We may have, uh, beautiful web user interface, uh, or things like that. But the, the truth is that behind Microsoft Office, we have millions of hours of, uh, qualified, uh, human labor and work, and we cannot just, uh, uh, create it with, by coding or however we call it now, right? Same with the contract.

AI may created one size fits all contracts. In my experience, that's the best way to end up in court. And, uh, as a practicing lawyer, what I see now, uh, steadily growing number of incoming clients saying, listen, we used AI to, you insert something to draft the contract to create an agreement to settle, and now we are being sued or we are willing to sue.

So, uh, amusingly, amusingly use of AI to fully replace lawyers, actually bring more work to lawyers. So I would suggest, uh, using ai, uh, if you're a lawyer, if you're an engineer, if you're a penetration tester, it'll certainly accelerate a lot of things. It'll certainly intelligently automate many, uh, time consuming, but not very, uh, exciting tasks and processes.

It'll boost your productivity, but it's not here to replace you. And last thing, probably, you know, from the technical viewpoint, we simply cannot have perfect ai. By perfect, I mean AI that will always produce secure, uh, privacy friendly and reliable code.

Why? Because AI was trained, I mean, all large language models, they were trained on, uh, different types of data, right? Including vulnerable code, including publicly available code, including open source software.

So all this code inevitably contains some vulnerabilities or even back doors. So when we have this kind of data in our training data set, inevitably sooner or later, our AI generated code will contain vulnerability or even a backdoor. So hypothetically, uh, we could create a perfect LLM model that would write, you know, flawless AI code or, or perfect, uh, from a legal viewpoint contracts.

But we don't have training data because in order to generate contracts, LLM models, uh, were trained on all kind of poorly written contracts. Uh, in order to create software, AI was trained on all kind of relatively, uh, insecurely written code. So we cannot really take it out.

And I believe that AI is great. AI is here to stay, AI is awesome, and I personally use AI for many different, uh, types of tasks, but it's not here to replace any of us all the time. Big, at least I don't disagree with you.

I, I, you know, I tell my people here, you, AI is not gonna take your job. Someone who uses AI better than you is gonna take your job. And, um, I, and I think it's particularly true, even in law, quite frankly, right?

I mean, yes, law was based on, you know, precedents and, and, and studied decisis and so forth, but, um, a good lawyer using AI will be a better lawyer. AI trying to be a lawyer is not a very good lawyer. Do, do you understand what I'm saying?

Absolutely. And, and, and I think the same thing is true for coding. I think the same thing is true for marketing.

I think the same thing is true for, you know, pick, pick a use case with ai. I think having that human, um, expertise and leveraging AI with it is, is at least for the, as you say, foreseeable future, I think is where we're going. Um, however, it is turning up a lot of vulnerabilities, but it's a funny thing about these vulnerabilities that it's finding, I don't wanna say they're not real because who's, who's to say what a real vulnerability is, right?

But it's, it, it doesn't distinguish between something that's exploitable reachable, you know, truly, you know, when we look at the criticality, whether you want to use CVE or something else, it is, I think it has a harder time distinguishing that, and that's still sort of the human purview there. But, you know, I, I don't wanna poo poo it either. That's a a, a very famous word there, poo poo.

Um, but I don't wanna poo pooh it in that it's, it's trivial. It's certainly not trivial, right? It, it is going to be a, a big part of it.

I guess my question to you then, IA, is as the CEO, what, what's immuno web's view on ai? Is it a tool you're using to help your customers? Is it something you're still looking at at this point?

How do you see it as the CEO of a company? How do you see it playing out for you? Uh, so basically our strategy, uh, has been consistent since 2019.

So we've been always transparently telling our customers that list. And we, uh, use machine learning and artificial intelligence for intelligent automation and acceleration of different tasks and processes. For example, web application causing crawling, uh, vulnerability verification, many other things.

Uh, but AI is no magic. And when we talk about penetration testing, we always say, listen, uh, practically, and also from a legal viewpoint, you cannot have fully automated or AI powered, AI enabled penetration testing. You still need a human expert behind the wheel.

And, uh, we've been always, uh, transparently saying, listen, yes, with our AI technology, you'll probably have, uh, better results compared to fully automated software in terms of quality and quantity of findings. But it's certainly not a replacement of a qualified human penetration tester. And if you need penetration testing, you still need human brain here.

Yeah, agreed. Look, I, I, I, I also think that at least for the short term future, we're gonna be very much in a sort of uncharted waters kinda way where, and you know, as a lawyer, the law is usually not a leading indicator. We're more of a trailing indicator.

It's gonna take time for the legal system to catch up with what is going on in the marketplace here, right? I mean, cases will have to work their way through the system. They'll have to be reasoned out and stuff like that.

Um, and, and so in this, in between time, if you will, we will see more uncertainty. We probably will see more lawsuits, incidents as, you know, as we kind of map out if you will, you know, where, where we're going. So it's gonna be interesting times.

I, we're almost out of time though. But for people who want to get more information on immuno web, where do you, where can they go? com and we have all the information transparently available there.

All right. Hey, come back on. We always talk about AI and how this is affecting us, so I'd love to have you back, come, maybe we could get you in with one of the panels we're doing on this.

'cause they're fascinating conversations. Lovely. Thank you so much.

Alright, Dr. IA Chenko, CEO Immuno Web here on Textron tv. We're gonna take a break.

We'll be back in a moment. Hi everybody. Thank you for joining us on this series about AI in the mainframe environment.

My name is Mitch Ashley, and I lead the software lifecycle engineering practice with the Futurum Group. Now, this is part of a three part series, segment number two, where we're gonna be discussing infusing intelligence using AI as a partner in in mainframe environment with mainframe teams. Today I am joined by Anthony as Starro.

Anthony is senior director of, uh, architecture for AI at BMC software. Great to be talking with you, Anthony. Thanks.

Pitch. Thanks for having me. You know, when organizations feel they're ready, kind of take that next step.

How does generative AI start to make an impact in the mainframe environment? Yeah, so, you know, we start looking at AI as a partner. Yeah, I was a developer for 30 plus years.

Boy, I would've loved to have AI at some point in my career. Why? Because from a developer's perspective, generative AI would've taken a lot of the toil out of my day to day activities.

I could have used AI in a way where just the grunt work I had to do day in and day out during my developer developer journey. AI could have been a tremendous help. I i, in that regard, take the friction outta outta my day.

When it comes to the AI ops space, as an example, I spent a lot of years building, uh, data visualization solutions, dashboards, et cetera. AI as a partner in that journey is how many times are we looking at operational dashboards and we're seeing blinking lights, or maybe we're not seeing certain things that may be in graphs or a data grid that's being shown. And, you know, we're, we're, we're looking at it.

It would be nice if we had AI sitting there as your partner observing the same thing you are, and then pointing things out to you, uh, that you would otherwise miss. So I really liked that portion of generative AI making an impact day in and day out. This is kind of tied to what we talked about previous, previously.

If I'm a next generation mainframer and I'm new to the mainframe space, having AI as a partner in my daily journey, whether that's infusion in products that I'm using, or that's knowledge based access, as we discussed, all of that constitutes that generative AI bubble to help me start my journey on that mainframe space. So that's really where I see it. Helping folks on the mainframe accelerate their day by taking things off their plate that they shouldn't be worried about, and focus on the more important high value items and innovation that they should be doing day in and day out.

You know, going from AI as being ready as an advisor now into AI as a partner, it seems like we're using ai, we're working alongside us as we're performing work, and it's providing information and, you know, giving us insights. Maybe as you mentioned that we didn't have, I'm curious if you have some kinda real world examples. Regenerative AI is helping us.

We do. I, I I, I, I mentioned I talk to a lot of customers, and one of the things that comes up all the time is call ball, uh, code explain. They're looking at, you know, they got that next generation looking at code bases, that a 25, 30 year plus.

And there's a lot of complexity. There's code comments in there that are outta date. There are module comments that are there in there that are outta date.

Real world example, we've delivered through VMC. Any assistant that capability to do that code explain to do that. Document generation customers are using that.

They've given us really good feedback on the usage with that. That's a high value return. Another area that we focused on was our AI ops space, where BMC Amy assistant was able to go in there and, and, and when there was a, a problem in the environment, what's the root cause of this problem?

We have a lot of very sophisticated machine learning models and information that's presented to the user in that product experience. When generative AI came onto the scene, it was a great opportunity for us is we were able now to take something that was very complex, to explain and describe in the user experience, and have BMC AMY assistant come in and then just explain what the problem is, what the root cause was in plain terms. Not only that DMC Amy assistant that was also able to give next step recommendations on how to resolve or prevent that problem from happening again.

So these are real world problems that we've gotten feedback on, on how the AI truly helped elevate the business value that we're delivering out of our solutions. Now, you've talked about AI as a partner, giving us insights, helping us along, maybe triaging or diagnosing what issues might be. There's so much knowledge has gained in that process, but we're losing that knowledge with a lot of our workforce as they retire, move on, et cetera.

Talk about that knowledge loss and how AI can help us with that. Yeah, so the knowledge loss is real as, as we all know. And, you know, an AI system is only as good as its knowledge.

And what I mean by that is not the LLM knowledge, it's your institutional knowledge. It's the knowledge that your staff carries day in and day out. How do we capture that?

How do we infuse that into an AI system such that the AI system has more context and relevance and becomes smarter for the users using that? That is critical. So how do we capture that institutional knowledge?

We, we, we augment, or we have a facility within B-M-C-A-M-E assistant to capture that information and infuse it once it's infused. That's where the real power is. And it's not just a chat conversation where you're gonna go to BMC AMY assistant and you're gonna ask questions and you get responses back based on that captured knowledge, right?

We talked about that in the previous video that was around the advisor model. But, but imagine capturing knowledge and infusing in a, in a product experience, maybe it's in the AI ops space, capturing that knowledge so that when events happen and things happen within a dashboard or a report gets generated and there's alerts and exceptions and alarms that were generated, instead of just giving these generic type of, uh, insights out of the product, we can customize those responses and put them in terms that are relevant for that organization. And the way we could do that is by capturing that tribal knowledge.

I have to repeat that. I can't say tribal knowledge is capturing that enterprise knowledge that your senior staff has, and based on, uh, on processes and workflows they have when certain situations arise, we can infuse that in. So the AI can respond with, in a way that, that, that is relevant to your organization, for your organization to take the right steps that's relevant for you.

That's what capturing institutional knowledge within an AI experience to give you back responses in that regard. So, as we move along the journey that you've talked about, you know, adapting, adapting and using ai, going from the stage of giving advice as we're working to actually guiding work, um, I don't know that we want to turn it all over to the AI all at once, right? There's a step along the way, maybe a hybrid kind of AI talk about what that approach might look like.

Yeah, that, that's a, that's a really good question. So at this point, we've been talking about language models from the AI perspective and augmenting that with institutional knowledge or real time data as an example. But generative AI itself is not enough, right?

And when, when you, when you're looking at, um, infusing AI in, in, in your solutions and delivering that value to our customers, and a lot of times it's a, it, it, it's an, it's a, a, a collection of different AI techniques that we need to pull together to formulate the, the answer or the response that we want to give. And it's this combining of different AI techniques together to formulate that response. That's what we call hybrid ai.

So, as an example, I mentioned the BMC, uh, AMY Ops insight product. There's a lot of sophisticated machine learning models driving that product, that experience. But we augmented that with generative a AI for the explainability and the next step recommendations.

So the, it's, it's the combination use of machine learning models with generative AI language models working them together. That's a great example of a hybrid AI solution. So you can combine things like, you know, classic rules-based AI with language models or rules-based AI with, with machine learning models.

What, what, whatever the application may be. It's this combination together. That's the way we talk about it, is in the terms of hybrid ai.

And that's a key element on our next video on how our AI agents think along their journey. Very good. That'll be a nice incentive to watch that third part, I'm sure.

Yes. Well, before we get there, any, any final advice you have for organizations as they're beginning to integrate generative ai? So when it comes to generative ai, um, again, you, you gotta be really practical on your use cases that you want to apply with, uh, generative ai.

Start with the explainability, no matter what your area is or, uh, your domain is that you wanna, uh, uh, apply ai, start with generative AI with explainability. Once you get that sorted out, move on to that next step. That next step is when you explain the situation, what's happening here?

How do we do the next step? What's my next step actions? Whether that's to resolve a problem or the next step action could be very, could be simply as notifying someone that a certain situation is going on.

But take those journey, that type of journey with your ai, do the, do the explain part, and then move to recommendations and next step with, with generative ai. And how you're gonna do that is, especially with the recommend part, that's where you capture that institutional knowledge that you have with your senior staff that you infused into your AI system. It could provide all those insights to the AI I AI system on guiding it towards tho that output on those next steps.

Well, thank you very much, Anthony. It's great for you to share, uh, your experiences of your journey that you can help others along the way in that process. So you, you've joined us for our second segment in this three part series talking about infusing intelligence AI as a partner, generative AI in the main mainframe environment.

We're sure happy that you've joined us, and we hope that you'll stick around and, uh, check out the third segment. We'll kind of give you a little hint there of some of the things that we're gonna be talking about. It's about acting with confidence with, we have AI as an agent of change that's making things happen for us and with us in this environment.

Thanks again for joining us with you on the next segment. Hello and welcome to the latest edition of the Techstrong AI Leadership Insight series. I'm your host, Mike Vizard today with Ben Schreiner, who's head of AI and modern data strategy business development for AWS.

And while we're having a little chat about how small businesses are using ai, Ben, welcome to show. Thanks. Happy to be here.

I think we've all been heavily focused on larger enterprises 'cause well, they have enough money to go and experiment with different things, but I think smaller businesses tend to be a little more careful about what they invest in and what they're gonna be willing to spend some time with. Yet, I know that they're playing around with ai, but from what you're seeing, is there a different set of behaviors and patterns of to how they approach ai? Um, I, I think your characterization is fair.

Um, they're definitely more, um, cautious with their money and time. Uh, I've yet to meet a small business owner, uh, who had enough time to do everything that they're, they're trying to do. Uh, but what we are seeing is, is them experiment, uh, early days with ai, um, in an effort to, to save time.

Um, AI has an amazing ability to, to help folks compress time in a lot of different areas. Um, and I'll say this about small and medium businesses, most will will leverage, you know, some of the free, you know, consumer based tools as their entry point, uh, you know, on their phone and, you know, helping 'em, you know, think through things. Um, but the other place that small, medium businesses are gonna see AI is in the software that they use.

And so, uh, here at AWS we're helping a number of software providers embed AI into their software that then gets consumed, you know, by a, a small, a small business, for example, uh, QuickBooks or TurboTax. Uh, we've been working with Intuit and, uh, uh, and the teams over there to, uh, enhance their software and, and embed ai. Do you think that that is maybe the primary vector through which most people get exposed to ai?

It starts showing up as some sort of AI agent embedded in a SaaS application that happens to be running on a cloud platform somewhere. But ultimately this is the easiest path of least resistance, shall we say? Yeah, I think, uh, again, most people will get exposed as, as AI, as a feature of the Software that they already Use.

Ideally, making that software easier to use, easier to interact with all of the promises of using natural language to maybe get, uh, new and different analysis or reports, uh, out of their accounting system, for example, or inventory management. Um, but where I see small businesses starting to get curious is, is truly on operational efficiency, right? And, and where or how, you know, they can better market, uh, or create, you know, um, yeah, let's call it, uh, you know, um, production level or, uh, um, agency level advertisements and, and imagery that maybe they couldn't afford in the past.

But now with some of these AI tools, they're able to develop some of that content, um, themselves without having to hire an outside agency and, and allowing them to do more themselves, uh, than maybe they did in, in the past. To your point about that, do small businesses have an advantage in terms of, well, there are just fewer silos in their organizations, and if I go into a large enterprise, sometimes getting multiple disciplines and groups to collaborate is challenging, whereas in a small business, well, necessity is often the mother of invention, right? It, it is.

I will say in, in all of my conversations with small and medium business owners and, and executives, if there is a compelling problem and there is a proven solution, they can move very quickly. And that is their competitive advantage in the marketplace. Uh, I would argue they're also able to adapt to changing customer, um, you know, needs and expectations, uh, a little faster, again, because they don't have, you know, maybe some of the bureaucracy or or bulk that a, a larger organization, uh, would have.

So that, that nimbleness, uh, is definitely something that, uh, many small and medium businesses wield, um, in order to gain market share and, uh, and, and, uh, uh, protect their competitive advantage. And I think AI presents an opportunity for them to do that again, um, and leverage the tools in new and different ways, um, and experiment. Um, but right now it's, it's really about learning what these tools can, can do, um, and how they can help.

And I, I, I give most, uh, executives this advice is to focus on real problems, right? Don't, just, don't just play with a new toy. Um, focus on a real problem.

Like, uh, where do you spend too much time? Or where are you losing money? How do you, how do you get ahead of it?

How do you become more predictive? How do you, um, have better, uh, visibility into your entire operation so you can focus on applying the technology in places where it can have the biggest impact, uh, solving real problems. That's where you're gonna get return on time and return on, on the invested money, um, that's associated with it.

And that's, that's the biggest benefit for any, uh, you know, leadership team or, or operator, is, is using these tools in places where it makes a difference. Do you also think that maybe they'll have a little more confidence in their ability to compete with larger companies? 'cause they'll be like, well, AI's leveling this playing field a little bit.

Uh, you know what I, I do think, uh, our, our efforts to democratize access to this, you know, transformational technology is gonna help, um, historically, uh, technology would be rolled out to only those who could afford it, right? The great big organizations, uh, would get it first and have, uh, you know, many years worth of competitive advantage before it trickled down, uh, to the small and medium business owners. Uh, this technology is very different.

And again, a w s's efforts to not only democratize access to the cloud, but also to, uh, to these transformational technologies like ai, um, as well as making multiple models available. Um, that's an area, again, of differentiation for us. Uh, we don't know, uh, you know, what model will work best for this use case or for that customer.

And so we make it available, um, a a marketplace where you can try different models to see what works best, evaluate them, um, and make the best decision for your organization. Versus, uh, being a, an organization that provides, uh, you know, we certainly have our own models, but we wanna make sure that, uh, our customers have a breadth of choice so they can make the right decisions for their business. Mm-hmm.

Will it get easier to orchestrate some of these AI agents? I think a lot of times folks get involved with various AI agents, especially in different SaaS apps, and they go, gee, I wish that AI agent would just talk to that other AI agent. And we can automate this even further.

I will say this, uh, again, if we're forward looking, we'll it get easier. Yes. Um, uh, at the same time, some things will get more complex, right?

And so, uh, you can imagine a world where, you know, the example you just gave, uh, I think a lot of people in their head want a personal assistant that helps them do the things that they need to get done in a, in a given day, uh, both professionally and, and, and potentially personally. Um, but there'll be a time where agents are interacting with each other. And then you have this concept of complexity because you could have hundreds or thousands or potentially millions of agents, um, and you need to be able to monitor and govern and ensure that things are being done in a secure way.

Um, and so, you know, it, it's gonna solve some problems. Um, it is definitely gonna create new problems. Um, that's the beauty of, of technology and, uh, the evolution of it, um, is we, we uncover new bottlenecks.

Um, but ultimately if we, if we're forward looking, we anticipate, uh, a need to monitor and govern a need to authenticate agents, uh, like Mike, your agent should have access to the data that you're allowed to have access to. My agent shouldn't be able to access that same data, right? If I'm, if I'm not entitled to it.

Like, uh, for example, the, the compensation for all the executives, right? There's only certain people that should have access to that, and the agents are a proxy for someone. They're doing something on someone's behalf.

And we need to make sure that that authentication and that security is maintained so that you're not accidentally leaking out, uh, you know, potentially, uh, you know, valuable information or, uh, you know, personally identifiable information or credit card information. Like, you wouldn't want these agents to be compromised and, and allowing, you know, somebody to be taking data out of your organization. So, um, you know, is this just an, it's a new paradigm, but some of our, you know, tried and true best practices when it comes to security need to be applied in this, in this new, uh, you know, brave new world, if you will.

Mm-hmm. Um, they say good fences make for good neighbors. So how do I put those boundaries in for AI agents and how do they get set up?

Yeah, I, uh, I've been advising folks who, who there was a, there was a rush to do ai, and there are a lot of folks that put, uh, you know, uh, uh, a chat bot, uh, you know, out, uh, as quickly as they could, um, and maybe skipped a few of the best practices on making sure, uh, security was set up appropriately. So I have a, I have a test that I ask folks to run if they have a, a corporate chat bot that they've deployed, is to ask it for a chocolate chip cookie recipe. Uh, because unless you're in the chocolate chip cookie making business, it shouldn't give you an answer, um, in, and I believe that we, the humans and the executives listening to this, um, it's our responsibility to put AI in a box and tell it what it can and can't do on our behalf.

Right? And if we don't do that, that's a, then we've invited risk to our organization that we probably don't want. Uh, and I used a cookie recipe as a, a, a pretty innocent question that you could ask.

Um, you could think of a lot more nefarious questions that you could ask AI to, to give you answers to. And if it does, uh, and then somebody goes and does something you know, bad, um, and it gets traced back to your corporation, uh, that's not what anybody wants. And so, uh, we have a responsibility, uh, to put appropriate guardrails around ai.

We also, as leaders, have a responsibility to our people, um, to bring them along for this journey to expose them and, and create an environment where they can learn how to use these tools to augment their productivity and their capabilities. So ultimately, you can serve more customers faster, better, uh, and more efficiently. Um, but ultimately, as, as a society, this is amazing transformational technology that I encourage everyone to learn as much as possible, uh, about it, figure out, even if you start out figuring out how it can help you personally, and then start to think about how you might be able to apply it, uh, inside your business.

Um, but these are, these are tools that we need to, to learn how to, how to leverage. Um, and the tools are only getting better. That's also the exciting part of, of what's going on here.

And you've been in technology for a while, as have I, and I've never seen anything move as fast as what we're seeing, uh, you know, over the last couple of years when it comes to AI evolution as well as investment. I sometimes feel like when I talk to people and there are issues that they've run into, usually can be traced back to, well, the AI agent was perhaps a little overly aggressive in its desire to be helpful. Yes.

And then they, it did something like the leading emails or whatever, right? Um, Is do we need to kind of revisit how these AI agents are kind of programmed because they are aggressively trying to help, but, you know, in some ways that can be the cure wines have been worse than the disease, right? Yeah.

I, unintended consequences we call it, right? Like, uh, and I, I do think right now, uh, even if we just look at, at, at prompting asking questions of a, of a, an ai, um, application, we are learning to ask better questions, right? Because if you ask not a great question, you get not a great answer.

And then you start to refine the questions and you get better and better answers. So we're providing more context to the agents so that they're able to do a better job, um, and understand what the intent is, have all the information they need to be successful, and that ensures a higher probability of success. Um, the generic agents that sort of you're describing that maybe haven't been given, um, you know, the right guardrails or parameters, um, with which to operate, we, we are learning, um, as we interact with these things and things go wrong, uh, where we need to put, uh, you know, better controls in place in order to protect the data, uh, or to make sure that something isn't, uh, it's not doing something we don't want it to do.

Um, you know, for example, you know, some of the agents that I've been working with, you know, I'll, it'll, it'll come up with a plan and then it asks me, it says, here's the plan. Do you want me to do it right? So it checks in with me before it goes and does it.

And that's a great check until the human gets a little lazy, right? And says yes. And then you find out something happened that you didn't want, and you go back in going, what happened?

And sure enough, it says it was gonna do it, and you just were quick to say yes because you just wanted to move on and probably didn't use your critical thinking or didn't read, uh, or pay attention to all the details. Um, and so I think having a human in the loop, um, and making sure that it's checking in with you, but that still puts the, the responsibility on us to make sure, again, that these, uh, these, uh, agents are doing what it is we want them to do. And I'd say, Mike, the important thing is, is when you start to have autonomous agents, you want a way to monitor, because there, there's this concept of drift where you're told it to do one thing, and then slowly over time, it's starts to drift off, uh, and do something else, or, or not exactly what you had said.

And you want a way to be able to notice when that's happening, right? And so maybe you set a, a pretty narrow, think of a, a, a a, a road and you have the lane and, and you wanna set the lane for ai, but you wanna know when it gets close to the edge, um, so that you can perhaps take corrective action, uh, to make sure it stays in the lane that you intend it to be in. And I think that's part of us learning how to deploy these things in ways where we can trust that it's doing what we want it to do and only what we want it to do.

And every time we have an example of something getting deleted or accidentally, you know, happening, it's an opportunity for us to learn how do we put those guardrails in place to make sure that that doesn't happen again? But it's a, it's a learning process, and I would, uh, I would submit to you, it doesn't matter how big the organization is, how much money they have, we're all still learning how to use these, uh, these tools and techniques. Um, that's some of the fun.

Um, and that's what I would encourage, again, all your listeners to, uh, to again, begin, begin the learning, um, learn as fast as you can. Um, what they're able to do is, is, is mind boggling. Like, like what we could do today, like fundamentally wasn't feasible two years ago.

And I'd argue what we're gonna be able to do in two years hasn't even been invented yet. And that's a, that's a pretty awe inspiring, you know, statement, um, to make. But I think your audience, the small and medium business owner has, has so much to gain if they figure out how to use these tools to their advantage.

Mm-hmm. I'm, you know, as I was listening to you, I couldn't help be reminded that, well, those interactions go on between humans all the time, right? We tell people we think we want 'em to do something, and then, you know, it winds up having those unintended consequences.

So sure does. Maybe the AI agents aren't that different, or is it just that the AI agents can't really undo something after they've done it the way and maybe humans can't? Um, you know what, again, I think this is an evolving area.

Um, one of the things I've been saying to folks, um, is that we, we are the last generation of managers to only manage people. Um, and so if you think about it for a second, I, I believe in the not too distant future, you'll have agents that are working on your behalf that you need to monitor like you would a human, and make sure that they're still doing what it is you want them to do. Uh, and you'll need to coach them and, and, and correct them, um, as needed so that they stay, you know, in that lane, very similar, uh, you know, to how you would train a a, a junior employee.

Um, there's a ramp time, uh, there's mistakes that are made, uh, and ideally both the human and AI can, can learn from those mistakes, um, and, and evolve and get better, uh, and become more valuable to you, uh, you know, and, and your organization. Um, ultimately, these AI agents will encounter AI agents that work for other businesses, and there'll be some level of negotiation that goes back and forth. And how do you envision that might happen?

'cause a small business might be trying to sell something at the highest margin, and the buyer of that service or product may be trying to buy it at the lowest margin they want. Those two agents just kind of try to beat each other up until they call for help. Uh, and that may be it, right?

That call for help becomes important. Um, we're already seeing agents be used to review contracts in terms of contracts, um, to analyze, uh, you know, service level agreements. Um, again, as, as you, as you grow in size, you have more and more of these, uh, binding relationships.

And, and certainly the scenario you just, um, you know, you know, gave where you have one agent representing one organization and another agent, uh, you know, maybe a buyer representing, you know, their organization. And, and as long as you've provided the parameters that say, you know, you can sell, I want you to optimize, you know, margin. And, uh, and, and here is the, the lane with which you can operate and provided, you know, those two things match up, and maybe it's a Venn diagram and there's a little bit of overlap, and they strike that balance.

But ultimately, if not right, in the scenario you gave where one agent's just trying to drive the price below where the other agent's willing to sell, then you have an impasse and potentially that gets escalated to say, Hey, humans, like we weren't able to come to an a mutually, you know, within our bounds, a mutually agreeable, um, you know, contract. Um, you know, do we want to ignore and move on? Or, or do you want to have an exception to the parameters that you've given me to, to operate?

And honestly, Mike, it's, it's no different than what we do with humans, right? You have a seller out there and they're given, uh, a margin target or, or don't go below this price. Um, but, you know, we're, we're, we're here to make money and, uh, you know, wanna obviously try and optimize our, our margins.

Um, and, uh, you know, I'm, I'm in the process of buying a car right now, and, and they've gone to their manager to see if they can accept my offer. Um, so it, it's, uh, it's this, it's the same parallel just in a, uh, uh, you know, potentially in a, a much broader, uh, you know, expansive context. And I would say this for this small, medium businesses, um, historically, we have thrown human capital as human capital as the, as the way to expand, right?

And, and what I want you to have this mental model is AI should be augmenting your humans so that they can do more, talk to more customers, serve more customers, but it becomes a scaling mechanism so that you don't just have to hire more and more people in order to scale your business. Um, I'm not saying you're not gonna have to hire people, um, that's not it at all. Um, but it is, uh, each human can now do, you know, maybe it's, you know, 30, 40% more.

And that allows you to ex to scale your business, um, and continue to invest, uh, both in your people and in your business' growth. Um, that's, that's one aspect. The other aspect I think AI brings to the table is tapping into human creativity.

Most small and medium businesses started because they saw an opportunity that wasn't being met by the big corporations, or they spun out of a big corporation to go do their own thing. And I think AI is gonna empower a whole bunch of new entrepreneurs to be able to take that risk, uh, and go and create value, uh, for their customers, uh, being able to create, uh, their own website, their own shop, um, and, and help customers in new and different ways, um, ultimately, um, you know, creating new products and services for all of us to enjoy. All right, folks, you heard it here in the next few years.

It may not be about the technology at all as much as just us humans and AI agents trying to figure out how to live with each other. Hey, Ben, thanks for being on the show. It was great to meet you, Mike.

Thanks for having me. All right. And thank you all for being watching.

Uh, try that again. And thank you all for watching the latest episode of the Techstrong AI Leadership Insight series. You can find this episode and others on our website.

We invite you to check them all out. Until then, we'll see you next. Welcome to Security Boulevard, the cybersecurity podcast from the Future Room Group.

Each episode explores a variety of topics within cybersecurity and the technologies that drive it. com, the Security Boulevard, YouTube Channel, tech Strong tv, and all of your favorite podcast platforms. I want you to meet today's, before we jump today, the episode starting with Fernando.

Fernando, it's good Old. So I, I'm, I'm a energy notd at a hundred percent. I'll, I'll do the best I can.

It's totally okay. Fernando, we, uh, you know how it is when it starts to get close to spring. Our body is like, what is this weather change?

Mm-hmm. Yes. And we have, and we're coming up with the time zone change, or not times, I'm sorry, daylight savings coming up next week.

Oh, yeah. By the time this episode comes out. We'll, or we'll be close if we haven't already done it, depending on which part of the world you're in.

com is your Friend. Yes, absolutely. And joining us also is Mitch Ashley.

Mitch, as always, good to see you as well. Great to see you. Great to be here with all of you.

Fernando, if I can give you some good news, you're getting, you're getting over the cold before RSAC instead of after. So look on the bright side, Let's hope. Yes, because yes.

So for so many reasons, Always, always, always great to be here, great to be talking, uh, cyber with you guys, enjoy it. Awesome. And of course, I'm Tom Hollingsworth, uh, all things related to security at Tech Field Day, which is a part of the Futurum Group.

But let's jump into today's episode. Now, when you listen to your favorite episode of Security Boulevard, and let's, let's face it, I hope they're all your favorite episodes. Uh, one of the things we talk about when we get to the end is we always talk about some of the research that our guests are doing as part of the Futurum group.

Well, today we wanted to talk about some of the research that we're doing here at the Futurum Group. And specifically, I wanted to let Mitch Ashley talk about this, because this is research that he has come up with that published, uh, about a week ago on February the 24th. So, Mitch, tell us what you have learned.

Well, you know, I, I wear this hat of having a lot of practitioner experience in my background. About two thirds of the time I've been like a CTO leading development teams in security products, as well as SaaS and other things. So I can't help, I can't help get away from some of that CTO, uh, DNA kind of the architecture of things.

And that, you know, I think what I struggle with in the industry with so much happening all at once is how do you put all the pieces together and say, what's happening? And is this that, or is this something else? And so what needs to happen?

And then also you combine, you know, Fernando and I are always talking about the things to do that will make our, make it easier for us to know that things are secured. And so, as I was doing our, our research, one of them, we do a buyer study. And that buyer study is people who are spending some money on software security.

Fernando one does on, does one on security, um, and a lot of different topics in that, uh, in my particular study, it asks about AI observable a lot of observability questions, but AI observability and, uh, agent observability, uh, AIOps, things like that. And in a one year period since I did the last, the second to last update to this, you know, there was one AI automation in the top 10 of what were the drivers for looking at, uh, ad observability platforms in this round, a little more than 12 months later, four are in the top 10, actually the top six, frankly. Um, AI observability, agent observability, uh, AI ops, and also cost optimization, including AI as well.

So the market is clearly recognized just as, as one set of data points that we need better observability, we need to be able to understand what is happening with ai. Now, let me back up for a minute, and then Fernando, I'd love to get you to, to chime into, so we're in this rush of recreating how we design and create software, right? The agents are part of that fabric where they are the fabric for a lot of the development, but we're also developing software with agents.

So the agents themselves are being used to create the agents that are the software that we wanna deploy into production. In the meantime, the enterprises are saying, wait, wait, wait. I need accountability for all this stuff.

I need governance. I need behavior analysis. I need guardrails.

I need governance. I need the ability to meet my compliance requirements. So before you deploy any of that at scale, I need you to address the security compliance, behavior issues that we need.

The visibility, you know, I think Fernando's heard me say, you know, trust is visibility. When you, when you experience it, you can see it, okay, I can trust it if I have, if it's trust me, you might trust me, but maybe that's not always a wise thing to do. So that's why I came up with this idea of observability native.

But before we get into that, that's sort of the long windup of where we're going and why this is observability is even elevated even more than it's been up to this point. Fernando, what's your Thoughts? No, I, I thank you.

I, I, it, it was interesting because, uh, we were just chatting after the episode before we reported, like I, I helped review the initial research and, and really liked it. I think that this comes with, uh, it's 2026. We're talking about AI all the time, right?

I think that this is a, a very useful frame to useful time to remind ourselves that specifically when it comes to security, right? And we'll, to the observability part in a second, but when it comes to security, we always talk about three major three ways to look at AI and security. AI for security, I'm taking AI capabilities, I'm deploying them for security use cases, ai, soc, ai, and the software development lifecycle, AI and GRC, AI and data classification.

Uh, and on and on and on. And like Tom and I are working together on something on sas, e, ai, and sass, e right? All those things, right?

The other one is, okay, security for ai. And, and that's the challenge, right? Okay.

You have your organization running, uh, uh, agents and ai, multiple scenarios. How do you ensure trust? And to your point, makes trust and visibility.

Absolutely. Right? But it's more than visibility, right?

Okay. How do you understand what the threat model is for that deployment of ai? Oh, look, I have an AI agent that is running on my machine, that is helping you with emails.

Oh, wonderful. Right? And, uh, what if the, the, how do you think through what bad things can happen with email, right?

And then things flow from there, right? But, but the notion of securing AI deployment, the third one, people don't talk about as much. I I think it's important.

It's a security from ai, right? It's okay, you have ai, even if you do nothing, right? Ai, uh, is being used by your adversaries, right?

Phishing, deep fakes and so on. What changes in your processes to handle this? But while these three major areas are, are, are important, right?

I think that the, the, the, the meta level problem at this is that we're asking, we, we have as security practitioners to embrace the, the, the non-technical side of us. We have to understand the business that our organizations do, right? Because the failures that we're seeing in, uh, in ai, they happen not just at the technical level.

They happen at the process level. They happen at the, and when I say process, I don't mean, uh, ps ef, i, I mean the business process, right? That too.

Business process too. But, but, but this is, but this is important because we're asking security teams and, and development teams and observability teams to better understand the processes around our business. Some of the most, uh, enjoyable time I spent in, in my past years has been on, on anti-fraud, right?

And it was fascinating to see, okay, how fraud actually happens. That that requires a level of understanding of what your, how your business operates. And I keep coming back to that as we need, we need to have that higher level of understanding of what our businesses are doing, right?

You can't just be a, uh, a security engineer who looks at the, at an, at an event on an endpoint, and okay, fine, it's, uh, it's raining, mimics minica is bad. Great, okay. Whatever.

Right? You need the higher level of understanding. Oh, look, this is actually a, uh, a workstation that an administrator used for, uh, that a domain admin used for something else.

And maybe that domain admin credential is there now. And by the way, that domain is actually a main corporate, uh, corporate. It's not just a test domain.

It's a, it's a live one. Anyway, the point being, I, I, uh, we need to understand what's going on. And that observability drive that you're talking about is, is, is really important.

No, I, I completely agree. And I think that one of the problems that we run into is that we have fallen into a world where observability is critical, but there's not as much transparency as we would like. Because one of the things that we're running into is a lot of the things, a lot of the context around what we're seeing out there is hidden from us.

Why do we do things the way that we do? Why does this policy exist? Uh, and and I think that from a business, you know, process perspective, you know, like, we do this because of regulation, or we do this because of that, but even some of the things that you wouldn't expect to, to filter through there need to be visible to organizations that are trying to build an automated strategy that has closed loop automation on top of it.

You know, we, as of this recording, if you have been in the news, you probably heard the horror story coming out of meta where an AI agent decided to delete somebody's entire inbox, even though you kept telling it to stop, it wouldn't stop. And you're thinking to yourself, well, who would delete an entire inbox? Because you have the kinds of observability to say, we don't just delete inboxes.

Like, even if we know it's an inbox we need to delete, we're more likely going to archive it for a period of time to make sure that it's not actually, it. There's nothing in there that we need. Or, you know, there's a lot of other things that we would do rather than just effectively removing it.

Or in, in the case of this agent, not just deleting the inbox, but deleting every message in the inbox one at a time, or in groups of 10, because well, why not? We didn't teach the agent not to do that. We, we hope that it had assumed that you shouldn't do that.

But that's, that's an observability transparency problem. What if corporate policy is, we don't delete inboxes, that we archive them, and then we, you know, remove them from the server at a point in time or whatever. Is that documented somewhere?

Is that written down somewhere so that the, the system knows that that's the way we do it. And if you're just sitting there shrugging your shoulders going, well, I don't know, well then why not? Like, because if you don't have that, you can't truly say that you have full observability.

If there's something happening in the system that can't be captured, that can't be documented, and it can't be audited, then you don't have observability. You have a, a hope and a prayer. And Well go ahead, Mitch.

I, I have something, but No, I, I would, I was just gonna say that in, in the, the world we're operating in, you know, we say in, in security, I, I must, I might need to know what we have, what's the inventory? What are all the things that we need to secure, right? Whether it's environments, software, networks, et cetera.

So we always do an inventory first. And the challenge is with ai, especially with agents, is, you know, in, in security in an ephemeral environment is a challenge, right? So that thing goes away.

I need to make sure I've got those, that information, kind of your inbox example, Tom, when it goes away, I wanna make sure I've got the information that I can get back to what was happening in, in that environment. Now we're talking about ephemeral environments where the code is ephemeral. AI is building code today in our development processes, but those agents can also create prompts in and of themselves.

They could actually write, write code themselves if we allowed them to do that. And of course, that's the question. The, the interesting word that comes out now is I need not just visibility for trust.

To your point, Fernando, I need accountability for trust. I need to know that end to end what the agent is doing or all those agents are doing. I can be accountable or they can be accountable.

Someone in the organization can know what it is and know that we've applied controls at all the right places and can verify that that's happening, even if it's a, an ephemeral environment. Okay? I dunno if I teed up what you're gonna say.

I, Fernando, I thank you for bringing up accountability. I, uh, it's, it's a huge problem. And we, we'll talk about that in a bit.

The thing I, I, I wanted to bring up is that we need, and, and thank you for, for bringing up the, the, the example. I was thinking about that as well, and, and what happened. So I think it's important.

This is the thing, I, I think it's important for us to understand how things work. So we understand how they break, to the best of my knowledge, the way what happened at, at, in the meta incident was that the, the instruction do not delete my inbox, fell out of the, the, the, the, the window for the agent fell out. So the agent didn't know and just kept doing it.

Right? Which means that as we are deploying these things, we must understand how they work, right? Uh, we can't just say, oh, look, an agent is doing something very well.

How does that agent actually work? What is the reasoning engine for that agent? How do I validate what that, uh, uh, what that uh, uh, reasoning process is like, right?

And that those are, those are, uh, I I'd say unanswered questions at this point, right? Because we are the, the, the, the, the, we are playing with the, with with, uh, the, not with fire. We're, we're playing for a scenario here where if we define the tasks constrained enough, current LMS under what their capabilities are, we'll be able to do them properly.

Right? Fine. Maybe we can do that.

Maybe we can do that within the context of, of, uh, uh, software life software development. Maybe we can do that within the context of security engineering, of, of, of soc alert triage, right? The challenge is what happens when we ask it to do things that is broader than what're supposed to do, right?

If I take in a human scenario, it's, it's, uh, uh, just because, uh, I'm, uh, uh, first year intern at my company is amazing at writing code, doesn't mean I have, I automatically give them, here's our software, um, here's our, our, our software architecture, uh, plans for the next three years. Please go implement it, right? We, we have to understand the capabilities, but your point account about accountability is also, okay, who is responsible when something fails?

And that is the thing that, uh, I seriously, I was thinking about this in the shower today, right? Uh, because we are in a because that's just who you are, because we're, um, the, the, the problem with, with, uh, that, that we're facing is that there is no recourse, right? If I make a mistake at work, if any of us make a mistake at work, right?

We have, uh, social capital, professional capital embedded into what we do every day, right? And, and there is a negative consequence for us. Like, if, if I make a mistake and, and, and I'm let go, I don't wanna do that.

So I am motivated to not make that mistake. Agents have no such motivation, right? And I think that's something we Oh, that, that drives me crazy.

But it's interesting that you bring that up. 'cause you, you talk about the intern example, and if I give an intern a task that's above them, what am I expecting? Am I expecting the intern to go out and, and get like 14 eBooks on whatever topic that is and spend hours or weeks of their time devouring it and then doing what I said?

Or do I expect someone who I know isn't trained to ask for help? Because I would expect in a situation like that, if I am overseeing someone, if I'm training someone, that they would come back to me and say, Hey, wait a minute, I'm not sure about this. Can you give me some guidance?

And you can say, okay, well I've taught you this and, and this is how you do it. Okay, now I'm learning. Right?

Whereas AI generally doesn't know what to do there. And one of the points that I think is important in, in Mitch's outline here is feedback loop actuality, right? You have to be able to get, to have a feedback loop, take actions based on the inputs that you give it.

But what happens when the feedback loop breaks? Because there's only two options, right? It either does something or it does, it's three options.

It does what it's supposed to do, it doesn't do what it's supposed to do, like it breaks or it does something it's not supposed to do. And that's what we're seeing, like in the, in the example that we keep bringing up with meta where, where the bot ripped through everyone's email, what would I want it to do if it breaks? Like, you know, if it halts, should it throw an email and go, Hey, I couldn't complete the task because you didn't gimme enough information, or because it's outside of the bounds of what I'm supposed to do.

That's what we expect the intern to do, right? Do nothing, break nothing, uh, versus, oh, well, hey, I know that you told me that you wanted me to do this thing, but what I actually did instead was something completely different that was unexpected. Well, why didn't you ask me first?

Well, because the loop's not coded to do that. And this kind of goes back to the problem that we've been having with LLMs for the last two years. What happens when they encounter incomplete information?

Do they tell you that they did, or do they make something up and fill in the blank? Because their directives, otherwise were just say something, make something, something up. Whatever sounds good.

Let's, let's kind of build on, uh, I think for Fernando, the word you were looking for was context window when things page outta a context window, right? Lose it. So lemme lemme introduce this into our context window of our discussion.

So, uh, and we mentioned the, the word observability native, and I'm not gonna go through the whole thing. There's a whole, uh, we'll provide a link to the, uh, page that describes it in more detail, but the, just as a working definition for, what do I mean by observability native? Well, it, it's not software, it's not a product, but it's an architectural pro approach when, think of it, a reference model where agent behavior, and I'll break this down, is captured and structured first, first class telemetry embedded throughout the entire software development process all the way into production, even at that point of when you're in design plan mode, using an AI agent to help you with that, all the way operating, this enabling, most importantly, machine speed governance and safe autonomous operation.

Okay? A lot of nice words in there. But essentially what that breaks down to is a couple of things.

One is, you hear me say the words first class telemetry. What I mean by that is it's not something we capture later. Let's see what spilled out on the operations room floor for data that we could go put some telemetry about.

Unfortunately, have we have some o hotel information? Is that gonna help us understand what these agents are doing? And did they go through any of those three paths?

And which one, Tom? So you referenced something called the loop, the feedback loop. So what I started with in defining this as well, you have to think about what agents do and how would you describe that.

So the, the kind of structure that I created is this four steps that it goes through. An agent looks at what are the, what's the intent, what do you tell it to do? What does it think it's supposed to do?

What's the goals that it's supposed to do? Um, second is reasoning. What choices does it make and resources that it used to make those choices?

Does it rely on what models did it do some procedural logic to create that? Why did it decide that was the path to go down? The third is the constraints.

What constraints were applied when it was considering the path that it might go down? Okay, well, I would normally go go talk to this model and, and feed it this prompt, but I'm not allowed to do that. I'm not allowed to put a, in a prompt that says, how do we delete everybody's email boxes that's out of guard.

That's, that's one of the guardrails or something like that. And of course, what was the outcome? What did it do?

You know, you're, to your point, Tom, what are those actions that it did, didn't failed, succeeded doing? So when you have those four steps, at least you have something you can now understand of what was it trying to do? How did it decide what it wanted to do?

Did we control it? That accountability part, and then the outcomes. And if you have observability for those four steps, now you can start to understand what agents to do.

Now that's easier said than done. In order to do that, we have to build that observability into agents as they're created and as the agents that are creating them. I also need that observability so we can provide full accountability.

'cause agents, building agents is just a supply chain issue. It's a software supply chain. I mean, if I infected the in agents that are building our software, well, I've just captured your supply chain.

Same, same thing. So there are seven other principles around this. One of them was around that ai, around the first class telemetry for agent across the lifecycle.

But I'll pause there. Um, but I think that describes pretty well. Tom, this is relatively new to you.

Uh, Fernando got to peer review this and provide a lot of input. Interested in your feedback or your questions, thoughts on this? I think it's important to do this in the cycle because so many people wanna skip to the fourth part, the outcome, right?

They don't think about all the other stuff that goes along with it. Think about the number of, of people who just put in like terrible prompts and then they're like, well, why did I get terrible results? Well, what was the intent like?

Like you have to define that. Like, think about people who don't write a rough draft of, uh, paper and then they get a bad grade and they're like, well, why not? Well, because it's a jumbled mess, right?

And then you've gotta think through the reasoning. Every school student out there that's listening to this, which I hope if we have a few more now knows, show your work. Why?

Because if I know where your work is, I can see where you went wrong. And we, we don't have to back all the way back up the decision tree. We can start right here.

Those things are so critical. What do I want to have happen? What is the intent?

And, and show me what you were thinking, right? Like, like you've got to have that and constraints. Like anyone who's ever taken a test, especially a practical test, knows that there are constraints.

I'm A-C-C-I-E and there are constraints all over that test. Make the output of this command look like that, but don't use this command. I'm doing that on purpose.

I'm trying to force you down a specific path of reasoning when I do. And I wanna see how you adapt to that. You've got to, you've gotta iterate through that.

That's why we call it a loop, right? And I think that that's so critical because people wanna leave out the verse three steps because they require me to do stuff. They require me to think, they require me to put my effort into the model.

And I just want magic to happen. Well, if you've ever played d and d, magic doesn't just happen. You have to spend all night reading the book and learning the words and preparing the ounce of gold that you have to do to cast the spell.

And then tomorrow it looks like magic. 'cause you spend all this work ahead of time making it happen And camp so you rest. So you could cast those Spells exactly.

Like people just don't understand that there's a lot of prep work that goes into this Yeah. To get the outcome that you want. So yeah, saying that from a deservability perspective, you've got to have this iterative loop that you go through in order to make sure that you're getting proper output.

Because I bet you if you spend more time on the first three and less time on the fourth, what you get out of the fourth is gonna be a lot closer to what you want And Don't know why you got it. Sorry, go ahead, Fernan. And the thing I liked about that, that feedback loop when, when first came across it Mitchell's work is this is independent of your role, right?

If you are a a, a, the prototypical, uh, year one intern, right? Figure out intent, figure out reasoning, figure out guardrails, and then outcomes, right? If you are a senior experience security engineer, uh, working with ai, figure out intent, figure out, figure out constraints, figure out outcomes, right?

So I think it's, it's, it's really interesting. The other thing I, uh, I I, I picked out on, on Mitch's work is he does a lot of work on observ observability, uh, observability for, we're speaking primarily to a security audience, right? But observability is huge in, in DevOps.

And uh, the thing that is interesting is that, um, how how close those those areas actually are and how to security and, and I've been, I've been watching this industry for years and there's so many, uh, so many scenarios where you have this, this coming together of security and observability, uh, security a lot. Like I, I, I think I'm on record saying I didn't, I never liked DevSecOps, uh, as a term because I think that the SEC should be implied, right? It's, it's like a in in inside DevOps, right?

And, and it's really interesting to see Mitch bring this, this, this framework together because this is something that applies well enough to, to the, the operations teams that may focus primarily on observability. But there's a ton of security insights here. And same thing for the security teams.

Sorry, Mitch. No, I I kind of liken it to, you know, the, the whole shift left where we've talked a lot about have we, have we succeeded? Uh, why didn't it succeed?

Why did it fail? I think this is more about 3D security. It isn't shifting anywhere.

It's like you need security everywhere. You need it through the process that you create the software, you need it through your, your security operation center. You need it in the ops, you need it in platform engineering.

So why not, you know, why not engineer it upfront into what we're doing? And that's what, I don't know any other way we can provide accountability about AI without putting those signals into the processes that we use to develop it. And in the software that we develop in that process, then we can potentially use those.

And that the decision loop cycle, that's, that's a really simple idea, but it captures sen the essentials, which is, I, you know, my, my my whole reasoning here was don't create anything too complicated. People will go blah, whatever that meant. Okay, I can understand that.

Now, there's seven other principles that go along with this observability native, we've kind of talked a lot about, about, um, many of them. I'm not gonna go in through each of them or even go into them much more detail. You can read that in the paper, in the research report.

But it's a lot about this unification, this feedback idea, the signals being first class. I think things that most people would say, I agree with that. I know we don't do that in full, but that's what we should be.

When I go to my observability vendor, I wanna ask them whether they use this term or not. Talk to me about observability native. How is observability built natively into what you're doing?

Whether it's security software, apps, uh, productivity tools, development tools. I wanna know that I can provide the accountability that my leadership and my industry, my regulators are gonna require of me. I I love the, the, the conversation that you just had in your head out loud with us, because it's every conversation I've ever had at every place I've ever worked.

Man, that's a really great idea. Well, then why aren't we doing it? I dunno, like, like is there a process?

Is there a thing that we do? And, and the, you know, to use the old Axo of, you know, when's the best time to write all that down? Well, when you started, well, when's the second best time?

How about right now? How having a a, a framework to work in is a great idea? Because it gives you checkpoints, it helps you figure out along the way, are we still doing what we're supposed to do?

And yeah, some things you feel are not useful to put into a framework or to have principles that you follow, you are wrong. Because if you don't, then you don't have a reason to justify the things that you do. And, and any kind of thing that's repeatable, that is, uh, capable of be being built upon has those things like, you know, we'll, we'll reference back to this, or We'll understand how this works.

Or, you know, is this something we should do? Well, let's follow one of these things. And Fernanda, to the point you brought up earlier about, you know, I feel like not having SEC on anything is kind of redundant.

'cause security should be integrated into everything. Boy, you'd think that, and yet we're still in a world where if you don't call out the fact that this needs to be secured, people won't secure it. Like, just, it, it's something that, well, other than security professionals like us who just assume that everything should be secured no matter what, we have a hard time with that because it's friction, right?

People don't want to be doing things slow them down, or that force them to second guess stuff until something deletes their whole inbox. Or someone hijacks their email and uses it to spam a thousand people. Like, like every, every one of these mistakes was written in a bad security policy somewhere because we let it happen.

Right? I I, but going back to what, you know, Mitch has, has written up here, like a lot of those things could have been stopped along the way if we'd have just followed some of these processes and applied them. And like you said, this is very focused on security.

'cause the name of the podcast is Security Boulevard. But a lot of this stuff works well everywhere. It does.

You know, just to share a little bit. 'cause in doing something like this, I asked myself, well, am I tilting at windmills? Is this another DevSecOps of, uh, should we should do this?

But it done never totally happens. And what inspired me to, to create it, part of it was working with folks like you in, in, in Fernando, just talking about security and depth about ai. Another part was also the in industry's reaction to ai.

What, because AI is different. It's non-deterministic. There's a lot of things that we don't control.

This whole accountability, this whole, how I'm not putting AI into production at scale without some things that we need in place. The markets responded. Splunk did.

Its, um, data security fabric. I'm not using the right name of it, but their, the, their, their data fabric for security, um, uh, DA Dynatrace and Datadog are all rushing into the market, but they're moving upstream. They wanna be part of the platform, part of GitHub's agent XQ or, uh, Amazon Bedrock Agent Core.

You hear these announcements over and over, or Google just, uh, updated their agent, uh, development kit, which is now really more of an execution framework than it is a, a development kit. But the announcements were all about, at least some of them were all about, uh, observability as well as development and planning and projects, et cetera, the whole life cycle. So the industry is moving this way, not 'cause Mitch, Mitch thought was a good idea.

'cause we all sat around and, you know, kind of complained, waxing poet, like, why doesn't it happen? Why isn't it happening? It's happening.

And the purpose of the framework is just give us a way to kind of understand is this, where does this fit in and what are they doing? And why does it matter? Yeah.

One of the things that, like, I, I go back to my soapbox of, of, of having a think broader. One of the things I really appreciate, what, what you've done is that it gives us, it, it gives us the grounding to start asking these questions, abstracting away some of the technology details. Like, okay, we, we don't need to talk about, we don't need to talk about, we don't need to talk.

Okay, we're thinking about this broader feedback loop. We're thinking about these constraints on, on the environment. Let's talk about those.

And then, so I think it's, it's particularly useful in, in that sense, I think that you have, uh, uh, uh, I, I go back to the fascination and the challenge that we have to deal with, with, with AI is this combination of understanding how it works so that we understand where it breaks, right? And then once we understand that, okay, how do we build our organization around it? It's funny that, um, as we track AI developments and whatnot, I see lots of people, I had to find economic references, right?

Tom, uh, uh, I see lots of people, um, making references to Conway's Law, right? Which is that principle of organization that what you ship resembles your organization. Organization, right?

Uh, and what AI is forcing us to do is rethinking what we should therefore, in a way, rethinking our organization. So how do we help security teams participate in that conversation about reshaping your organization? Uh, a little bit higher level stuff than usual, but I think it's absolutely important.

Well, I think we're gonna go Ahead and wrap it there because we've, we've talked a lot about this, but really what you should do is you should go check out what Mitch has written. Mitch, if people wanna find out more about the seven principles of Observability native, how can they do that? Easiest way is, um, we'll provide a link in the description, but you can also Google or Perplex, whatever you use, just, uh, Google futur and observability native.

It's not a term that's used a lot. Uh, well, hopefully it'll be over time, uh, but you'll find it really quickly. com/mitch-ashley, and there's post there.

But that way you don't have to kind of work through it. And I'm happy to direct people to it if you reach out to me on LinkedIn. Absolutely.

You should totally check it out, because there's a lot of great information there. Uh, Fernando, I know you're busy working on some stuff as well. What can people find out more about what you're doing?

Yep. No, I think that I'll, I'll point people to, to our main Foot Room website and, and whatnot. I think that, uh, as we're recording this, we just, uh, we just turned into March and, uh, which means that in, uh, in about 21 days, uh, I'll be at, uh, I'll be at RSAC alongside Mitch, alongside you.

So my brain is, is pretty much fried right now with, uh, what's going on at the RSAC. So that's, that's where my mind is right now, plus the, the, the, some of the faci work that you and I are doing and so on. So that's always fun too.

Look forward to seeing that, guys. Yeah. com about what we're doing around RSAC conference this year.

I just actually published the schedule for the event. You're gonna wanna make sure you check that out, uh, as well as all the other things that we have going on around observability and cloud and well security. Um, but we wanna thank you very much for listening to this podcast, and we hope that you enjoyed the conversation.

If you did, please make sure that you were subscribed on YouTube with notifications turned on, or that you are subscribed in your favorite podcast application, because we don't want you to miss any of the exciting episodes that we have coming up. You can also leave us a rating and a review and a comment. All of those things help us reach a wider audience of people that might be interested in security.

com in the Futureum group. com, text drunk tv website, or the Text Drunk TV app, which is available on all of the things, including all that new stuff that Apple is releasing this week, Security Boulevard on Twitter. See everyone.

Hey everyone, it's Alan Shimel, founder, CEO here at Techstrong, and welcome to our continuing series on, uh, AI agentic AI in the future here with, uh, the Microsoft team and our future of analyst team, as well as Techstrong. In this next episode, though, we're gonna be joined by Mitch Ashley, uh, of Futur, who leads the software development lifecycle and building segment at futurum. And Mitchell is talking with Brian Good, whose official titles is corporate Vice President and Agents marketing.

But Brian is really here talking about agent apps and chat, and it, it's, uh, you know, obviously a very hot topic as we move to an ent ai workflow based basis. So let's join Mitchell and Brian here with me, and it's great to have them both. My name is Mitch Ashley, and I'm VP of practice, lead of the software Lifecycle engineering practice at Futureum Research.

And, and Mitch, uh, my name is Brian. Good. Uh, I lead the business applications and agents team, uh, here at Microsoft.

Uh, let's start here. 2025 is described as kinda an inflection point for AI adoption. What do you think are the most significant changes that, uh, you've seen in organizations as they're using AI and agents in their businesses this year?

Well, I absolutely agree. 2025 is really an inflection point, and we'll sometimes describe it as the year that the Frontier Firm was born. And you might've heard us talk about the Frontier Firm before.

It's this idea of companies that are putting AI really at the heart of their business, and it's enabling them to do things like reinvent the way they engage with the customers or transform their business processes inside their company, um, and beyond. And it's really the year these frontier firms are sort of rising up and at time when I think we can learn a lot from these early adopters and understanding how they're deploying ai, how they're being successful, and then figure out how we can take those insights and bring 'em to, to our own businesses. That's where you can have outsized impact As AI transforms those functions.

What do you see as the new patterns of work that's enabled by copilot and agents as customers start leveraging the technology for productivity or innovation? I'll tell you what, I have studied these frontier firms as they, as they come up, and there's really three patterns that I see across these frontier firms. The first is really enabling, uh, employee productivity.

So they give every employee, uh, an AI assistant like Microsoft 365 copilot, and it helps them, those employees be more productive. The second pattern that I see is, uh, really these frontier firms deploying AI to automate existing business processes. So, for example, they already have a way of handling expense reports, but they can use AI to speed that up and reduce costs, and, and that certainly results in some benefits to the customer.

The third pattern is really where a customer, like starts from the beginning, let's say from first principles, and they reimagine a function altogether. They'll reimagine what it means to engage with a customer who has a, uh, an issue with their product, and they'll put agents at the heart of that. Most companies can only take on one or two of these functional transformation projects at, at any given time because it's a big lift.

Again, it's reimagining a function from first principles. It's not just taking existing processes and and applying AI to them. How far along do you think most organizations are in their adoption cycle for generative ai?

Well, I think we're still in early innings, uh, that there's no doubt about that. Um, you know, customers are starting to deploy AI in different functions as we talked about, but certainly they haven't, in most cases, fully realized kind of the transformation that AI can have across their organization. So it's still very early innings, but I'd say we see very promising signs and, and green shoots, uh, of that, uh, of that adoption and success.

Uh, again, the key really here is to start function by function, think about a particular function, think about peeling it back to the business processes you need to go focus on. That's where you can have outsized impact. How do you define AgTech business applications, and why are they critical for organizations?

Yeah. Well, I do have a vision that there's, uh, the application of old is really transformed with ai, and we call that new type of application, the AG agentic business application. And the ag agentic biz app includes the assistant for the human to start to use.

It includes prebuilt business process agents that basically take the drudgery out of work, and then it's built on a data foundation. And so instead of being limited just to the data that, you know, maybe is in the CRM system or the ERP system, it joins that with data from other lines of business systems and even productivity data so that you have a rich set of data that you can build agents on top of, and you can empower your humans to get better decisions from. And so this idea that brings all those together, the assistant, the agent, the application, I call the agentic vis app, and I think it's a really big idea.

Let's talk about why you're optimistic about the future of ag agentic business applications. I have a lot of optimism here because number one, I see customers already deploying these applications, uh, and and really transforming their business. So I already see people starting to get great benefit from it, but at a more human level, the thing that gets me excited is like, if you think about every one of our jobs, like there's a lot of stuff that we end up having to do that really isn't adding joy to our lives.

Uh, you know, me doing an expense report or, you know, filling out a CRM uh, system, you know, with an update from a customer call, those aren't the things that, uh, that make us, uh, unique. Those aren't the things that bring joy. Those aren't really even things that add business value collectively.

But if we can delegate those things to AI and agents, I think we'll enable ourselves to actually go do far bigger things and really take our ambition to the next level. And so I am absolutely excited for what the future holds. Yeah, I'd love to hear about how you see the role of AI agents evolving just over the next few years, especially as organizations start to redesign core business processes drive outcomes for efficiency.

How, how do you see this taking shape? Yeah, great question. You know, I, I, I would say that we really believe there's a spectrum, uh, of, of agents.

You know, there will be very simple agents that someone will use that might just be grounded on a particular knowledge source and help you, you know, understand, you know, what, what might happen from that, uh, from that knowledge source. Then there'll be task-based agents, and then they'll also be more sophisticated, fully autonomous agents as well. And this more autonomous agents is where you can unlock a lot of business value.

These are agents that, you know, aren't called by a by a human. They're just running independently. They're triggered based on different actions, and they can really automate business processes in some cases from start to finish.

So we believe there's this spectrum of agents, and today, I'd say most ENT use cases start with those more simple kind of knowledge agents. Uh, but increasingly we're seeing customers bring in these task-based agents and autonomous agents, uh, to automate, uh, things, uh, end to end. How about the C-suite leaders?

How should they be thinking about investing in agent technologies for long-term value? Well, every C-suite leader I talk to today is already in on ai. Like, every one of them recognizes that it's a competitive advantage if they can move quickly, and if they don't move quickly, they recognize it could be a, a disruptive force in their industry.

But let's face it, ai, it's transforming businesses, it's transforming functions. It's, it's gonna reshape entire industries. And every C-suite leader I talk to recognizes that and wants on board.

What they're looking for though, is a partner. They're looking for the tech, of course, they wanna make sure they've got the right tech, but they're also looking for a partner to help them shape this in their business. And that's where Microsoft, I think, comes into play.

Uh, you know, we're not a large scale model maker, uh, but we do take the best of the models and bring 'em into the workplace, uh, so that companies can use them. You know, we talk about AI being a disruptive technology, probably because it seems that it really can and will affect every part of our work, our lives, et cetera, certainly is affecting how we create software with new concept like agents and agentic ai. Curious about your thoughts.

Share with us, how does Microsoft view what the transformation is going to be like with AI really having an impact and a big benefit to businesses? Yeah. Many industry, uh, pundits will say that this move to AI agents is gonna, uh, lead to the rise of, uh, uh, more consumption like models or outcome-based pricing.

And I think they're right. I, that's definitely a direction that I see the world shifting as well. Um, the only thing I would, uh, balance that with is that many customers are still, uh, you know, most comfortable buying things on a, a per user or pre per seat basis.

And so, uh, the approach I'm taking is, you know, how do I enable customers to buy offerings that they're comfortable with? And that's typically like a per user or some type of per tenant type of type of license, while giving them the flexibility to, to grow and shift into more consumption models as they, uh, as their business changes. And so, uh, we are at an inflection point, just as we said, and I think one of the things that will change is the business model over time.

Talk some more about ag agentic when you think about agents operating on their own or more autonomously, and why, why is that an important thing that organizations are looking to move to? Yeah, it really comes down to business priorities. Like every business leader I talk to wants to find ways to grow their top line revenue, or they're looking for ways to automate, uh, things that, that they're doing so that they can save money or redirect folks to, uh, focus on more important activities.

And, you know, autonomous agents really fit that bill. You know, imagine in the sales context, you can have an autonomous agent, you know, going through marketing leads and qualifying them before handing them off to a human seller. That's work that wouldn't have gotten done in the past or would've been done by a human seller, uh, and have been relatively low value, not something they, they really enjoyed, uh, about their job.

And so an AI agent can do that and add great value to the company, and again, help that company grow on the top line. Talk About Microsoft, uh, and the offering that you have in the context of an end-to-end tool toolkit, if you will. Yeah.

A functional transformation. You know, as far as the toolkit goes, we really believe that there's three essential parts to it. The first is we think every employee should have an AI assistant in our case, uh, that's Microsoft 365 co-pilot, uh, think of that as a productivity tool to help every employee work get through their workday and get more done.

That can be quite transformative. That next step up is where agents come into the picture, and I describe agents as really being for every business process or workflow in an organization. And we have, uh, a toolkit that enables customers to build their own agents.

It starts with copilot studio, but even extends into our Azure capabilities with our, our Azure AI Foundry product. So agents pair very nicely with that co-pilot that I described first, and then the final step is where the system of record becomes the system of action. And we'll sometimes call this the agentic business application, where you take a CRM system and you add agents and an assistant to it to really transform those three are the essential ingredients or building blocks for AI transformation, um, in a frontier firm or any business at this point.

That's a great term. Can you share some examples of how Microsoft customers are already seeing measurable impact from adopting agent business applications? Yeah, one example I I think I can give is lifetime.

Uh, they're a great customer of ours, uh, based here in the United States. They've used us as part of their finance and supply chain operations, and they've, uh, deployed agents to basically speed up how they handle and process e-commerce orders. In fact, I think it saved them 95%, uh, in terms of their order, e-commerce, order efficiency by deploying, uh, AI agents and agentic business applications to solve that problem.

So I think that's a great example. We also have, uh, another great example, uh, from Europe, um, uh, a large utility named Enco who deployed a multi-language, uh, AI agent, uh, for their customers to help them scale and address customer questions. Uh, it's a pretty cool solution, saves them time and again, helps them scale up.

Now, it's a, it's a really exciting time in the world of agents. You know, you talked about the C-suite and the kind of the three phases in this adoption curve as we adopt ai. What, what kinda recommendations do you have?

Like, how to get started? Well, maybe near some of the near term activities. Well, you know, as I said earlier, I think AI is gonna transform every company, every function, every industry.

And that opportunity is so vast, sometimes it's hard to know where to get started. And I've got two bits of advice really there to, to anybody that, uh, that is, is pondering that question. Uh, the first thing is, again, start with a function.

Pick a function that you want to go after and then peel it like an onion. You know, go look at the next layer, which are the business processes in that function that you can apply, uh, a co-pilot or agents to, to really transform. The other thing though is like, don't get into analysis paralysis.

Just pick a business process. Just go pick a business process to get started with. And as you learn from applying AI to that business process, whatever it is, whatever your authority is, business process is, go apply AI to it.

You are gonna learn, your organization's gonna learn, your culture will adapt, and then it will flow from there. So the opportunity is so vast. Don't let that keep you from getting started.

You gotta get started, start simple, pick one thing and go from there. So as we move to an agentic business environment, let's talk about the people. How do you see the role of human creativity, judgment, leadership?

How is that gonna evolve? Yeah, well, that's an excellent question, and one of the things that we'll often talk about is how AI and frontier firms is gonna transform the way we work. It's gonna change, uh, the org chart into more of a work chart.

Uh, we sometimes talk about, uh, frontier firms taking on the Hollywood model where individuals will swarm around a problem, like focus on a problem and then disband, uh, you know, uh, when the, you know, once they've got a solution. And so it's absolutely gonna change the way we work with others inside the workplace. It also, I think, is gonna give rise to a new idea that we call an agent boss.

And you can imagine, just as a people manager today might take work and delegate it to different humans on their team, uh, you know, resolve conflicts and sort of manage performance. Every one of us in the future is gonna do that, but not just with humans, but also with agents. So imagine taking a business problem, breaking it up into pieces, delegating it to agents or agents, uh, on your team, resolving conflicts that might come up, applying human judgment.

Uh, it's gonna be an absolutely transformational moment, and I'm really excited about it. I, I really believe that, you know, there is still a huge opportunity for humans with human ambition to go do great work amplified by ai. Talk a little bit about how you see the difference in, in the working together between applications, assistance agents, the different technologies.

Well, that's an excellent question. And you know, we really have this complete toolkit that spans everything from the assistant to the agent, to the application. And I think those three things work together in a symbiotic way.

As an example, you can imagine that I might go to my assistant and, you know, ask a simple question, my AI assistant like Microsoft 365 co-pilot and ask a question about, you know, uh, summarize the emails or help me respond to the emails I've got. Or I might use an agent to update a CRM record after I meet with a customer, but I'm still gonna want to go to an application. Really, that's a purpose-built experience for me if I want a more specialized or fine tuned experience.

So those three things really work together, Like a little bit about the industries or maybe the business functions that you expect to see reshaped first Biogen AI transformation. Yeah, there are three or four, I'd say functions in particular that are, I'd say, um, you know, ground zero for, uh, functional, uh, transformation with AI and agents. Certainly customer service and customer experience is one that is, uh, absolutely being reshaped.

And I'd say a very early adopter of ai, no doubt about it. Another, where I'm seeing a lot of early AI adoption, uh, is in sales. Uh, and it's just because, just as we talked about, there's a big opportunity to use AI to basically increase capacity for that organization to grow top line revenue.

So the business impact there is undeniable, but I also see it in places like finance and supply chain, where you can use AI to shorten the time it takes to, from an order to actually being able to ship it that order. We're seeing people use AI to improve accounts, uh, payable and accounts receivable. Um, so we're seeing some pretty, uh, interesting impacts there.

Let's talk a little bit more about the frontier firms. You know, they're not just adopting technology, they're also changing the way they're do doing business around AI agents and co violet capabilities. You talk about, you know, what they're doing to invest and support the short term ROI that they were looking to get for long-term innovation.

The most successful frontier firms are doing actually is taking a functional approach. And so certainly they'll think about their entire company, uh, but they'll really take an approach that's function by function. They'll think about their sales function as an example, and then they'll peel back the onion a little bit and understand which processes inside their sales department they can automate.

Uh, using AI agents as an example. They'll look at which, uh, things their salespeople need help with, where you could pair up an assistant like co-pilot to help them throughout their workday. And they'll even look at how they can bring agents in to really augment business capacity, maybe to grow top line revenue or take out costs.

But starting function by function has really been the recipe that these, uh, frontier firms, uh, are using to see success. As an example, in our sales organization, uh, by deploying co-pilot and agents, we've been able to improve revenue per seller in some cases by almost 10% in some organizations. And, you know, if you think about that giving a seller 10% more capacity is like giving them an additional month, uh, in a year, uh, without actually having them spend any extra hours through the work week.

And so there's some pretty remarkable results. That's just sales. We've been able to do the same in customer service and our finance department and our IT department, our legal team has been able to reduce costs by 5%, uh, by deploying AI and agents, uh, within their, uh, within their functions.

How do we ensure trust and transparency as ag agentic systems become more and more autonomous? Yeah, well, there's two things that we want to do to help with trust and transparency. The first is we have a rigorous set of principles on, uh, responsible ai.

So for any AI that, uh, Microsoft deploys, we adhere to a an important set of guidelines. Uh, and that's really table stakes. So that's the first piece, uh, responsible AI and our focus there.

The second thing that I think is important is we now have an opportunity to start to quantify the value and the impact that many agents will have. And we often will call that in the industry, we'll call that evals or benchmarks. And increasingly, I think we have an opportunity to help our customers understand how agents are being effective in their workplace using these evals and benchmarks on real world problems.

So for example, uh, we, uh, uh, a typical, uh, workflow will be a sales leader doing sales research to try to understand like how they might want to organize accounts or territories or, you know, reshape planning as they think about the year ahead. We recently released a new benchmark, uh, that we call the sales research bench, and it shows how agents can actually help sales leaders in that very specific job, and it's quantified. Uh, and so I think you'll start to see more of that.

And between following responsible AI standards and then using quantitative measures like evals and benchmarks to understand efficacy, I think we're really on the cusp of helping customers know how they can deploy AI in a safe way for real business results. Well, thank you, Brian, for sharing with us your insights and kinda look into the future, what's happening with the Gentech business applications? Thank you very much.

It's really amazing the pace at which AI has been adopted and continues to evolve. The technology evolves on a near daily basis, but at the same time, organizations have to figure out how they're gonna implement their AI strategies and what the business outcomes that they're most important to their business. I think it's very fascinating how Microsoft has approached the market, both from the standpoint of addressing the individual and their productivity, but also thinking about new workflows, new models of business, but doing that with not a set of tools, but a set of capabilities that provide integration with data process, workflow, and agentic ai.

No one knows for sure what the future holds and what an agentic business might really, really look like. But in situations like today, when we remove constraints of what we can do with technology thanks to ai, that's where the possibilities are created. You know, using tools like copilot, using applications like Dynamics 365 and we'll, we'll see what end users as well as technologists bring to bear in their ideas and how they reshape businesses day and tomorrow.

And how about you? But I'm super excited about the future that we're creating together thanks to working with technology companies and with end users like yourself. Hi everyone, my name's Kurt Kine.

Uh, I am senior director of AI marketing for Hammer Space. Um, and joining me on the call today, Sam, if you wanna introduce yourself. Good morning, Sam Newham.

I, uh, run our AI solutions practice here at Hammer Space. We appreciate the time today. Yeah, really excited to be here with everyone here in the room and online as well.

Um, here's a brief agenda. Uh, so I'll be going over a quick recap of just what is Hammer Space, what is the hammer space and what is Hammer space, um, and how we do what we do, because it really takes a little bit of time to understand how we're differentiated from the typical storage company out there. Um, we see ourselves much more as a data company.

And so, um, it's an important difference, especially with today's workload demands and, um, infrastructure setups. Uh, then I'll kick it over to Sam and he'll be covering our Nvidia AI data platform integration and why that extends what Hammer space does into a whole new arena for folks who are deploying AI factories, um, and are looking to go from data chaos to AI ready data. And then finally, um, I will give an update on the open flash platform.

Uh, we talked to the delegates at a previous field day, um, about OFP and the initiative behind that, and we want to give you an update on practice and rare for a software company like Hammer Space, I'll even have some hardware to feely touchy. Yeah, it's exciting. Uh, and I'll be joined by, um, two folks, uh, Ted and John from Hammer Space to talk about why they're a key part of the initiative and our, uh, first reference to design.

Alright, so, so, uh, the hammer space, probably not so much for the people in the room, maybe a few people in the room haven't gotten this, uh, background before, but what is a hammer space, right? And it is that infinite space that you see in an animated show where somebody's carrying a bag and then they maybe pick, pull a giant hammer out of nowhere, right? It's this instantly accessible virtual space that is infinite in size.

And so that's a pretty good analogy for what we do for our customers, right? We disaggregate the data from the underlying infrastructure and allow that data to be in any location, in any cloud, on any infrastructure backend, and eliminate the silos that are so common in today's enterprises. And we accelerate pipelines by a aggregating the metadata that is attached to all of that data across the data state, making, you know, the infrastructure more valuable because we can continue to use whatever data you have in place and continue to use those systems, but then also rapidly deploy newer, cheaper, faster technology, integrate those really easily into your environment.

And with AI being such a big part of where we're generating data today, um, you know, the need for this massive repository that sits across, um, locations, clouds, and um, systems is really, really key. So a little bit of, you know, background on ai, right? I don't think this is a question too much anymore.

Um, but I think we often focus as vendors on our really cool system, right? And we don't focus so much, especially in the storage space on just how are people dealing with their data, right? And AI has really driven us as storage players to think differently, and we're not alone in that, um, at Hammer space.

But, you know, really going from where we were just serving files out, serving objects out into being able to serve something that's really useful for these AI applications, right? We're not just repositories anymore. Now we are really, you know, full data management systems and what hammer space does, is go beyond the system's view, right?

And we want to give people visibility into all of their data, not just the data sitting on one system. We want to be able to give people access to all the data across locations, give people access to all the data and visibility across their clouds. And we need to be able to automate those data pipelines to provide AI ready data, not just raw data out the front end.

And we need to be able to do that in real time. And so I've covered these problem areas, but I'll, I'll dive into each of them a little bit more, right? So, storage folks have been deploying systems for all sorts of rational reasons all over the place, right?

And each system might have a unique characteristic, right? You might deploy object storage over here because you have this data repository that you know is gonna grow and you don't want it to, you know, have onerous overhead. Uh, you want it to be easy to deploy, scale out.

Then you have NAS systems that are scale up maybe over on another side because you had more, uh, enterprise, you know, requirements for data protection, um, and things like that. And you wanted it easy to share and easy to access for folks, um, that didn't know anything about S3. Uh, then you right, started deploying pieces in the cloud because there was this infinite bucket in the cloud that allowed you to ship data up there.

But each of those new systems introduce complexity, additional data management overhead, um, you know, all the things that comes along with distributed data. Um, si similarly, all right, you stand up a western region, uh, you've got another office in East Coast, you've got another office in Europe, another office in Asia. All of these disconnected sites and need to be able to collaborate, especially now that we're applying AI applications to be able to gain better insights into, you know, everything that's spread out across the globe.

So how do you unify that without massive migrations and multiple copies spread all over the place? Additionally, we encounter more and more customers. Yes, Ray Ese, Silverton Consulting.

Um, you mentioned that you disaggregate data, but you aggregate metadata. You want kind of, I will go, I will dive deeper into that for sure. Okay.

Um, but essentially, right, that is hammer space is superpower, right? So we assimilate the metadata from the underlying storage system. So whether it's an object storage or a NASA system, um, from other system vendors out there.

And then we have metadata servers that sit outside the data path that manage all of the data across those systems and aggregate it into a single global namespace. And then we can have data writing to those systems. So you can leave the data in place, or you can move the data based on where it needs to be, when it needs to be there.

Um, and you can continue to utilize that underlying system under hammer spaces ownership. Uh, That assimilation process happens instantaneously, or, Uh, it's of course not instantaneously. It's, it's a days, it's a days versus month thing.

So when you look at, you know, what is a typical system vendor gonna come in and tell you is, you know, what, how are we gonna solve this problem? Well, we have the best system, we have the best new silo that's gonna solve all your silo problems. So just deploy our system, migrate all your data into our system, and then you have a single global namespace.

Instead, we say, leave all your data in place until you need to move it. So yes, there are cases where you're gonna have to move your data, and that's, you know, physics, but you don't have to move it all instantaneously. So that's a nice thing that when Hammer space comes in, you know, we're talking days to be able to stand up a global namespace as opposed to a gigantic migration process into our system.

And so you get access to this global namespace really, really quickly. And I talk about that a little bit later in a couple of customer use cases too. Okay?

Um, so, you know, too many storage systems to manage too many things out there. Um, none of them meet all the needs, right? I mean, as we're approaching, um, you know, infrastructure modernization, I think what is everybody trying to get to?

And that's a single converged infrastructure. You need to be able to serve S3, you need to be able to serve, um, a parallel file system. And then also, you know, N-F-S-S-M-B-C-S-I, um, MCP, all of these things need to come from a single system.

And we do all of that natively. And then we're very, very cloud friendly. We were built a software with a cloud first mentality, so that to be able to, rather than viewing cloud as a destination where you just spill bits out and then hope that you never need to bring 'em back because that's a really expensive product process.

You know, use the cloud as an extension of your on-prem infrastructure, right? And that speaks to, you know, the whole first to cloud type of, um, activity. And we can really help customers with an on-ramp to hybrid cloud settings.

And then finally, data management across systems has often been another new tool that you need to bring in. And so instead, you know, we bring data management into the storage layer and help people automate that via orchestration. And I'll go into that here in just a second.

Is Marian Newsom? I have a quick question. Yeah.

So you talked about aggregating the data from the metadata. How do you detect the drift if there's any between those two? Um, I will kick that one over to Sam, Sam if you're available for quick.

Yeah. The, the idea is to prevent data drift, right? It's one of the most poison things that happens in ai.

So by owning the metadata, right, by separating that particular thing, we synchronize those changes real time, right? Across our global namespace. And so we don't have some of that acknowledgement wait for this to write type situation.

And as part of our secret sauce is, is how we can resolve some of those discrepancies in multi writer situations and those types of things as well. But our system was tuned to deal with this first studio in the sky where we're doing multi editing, right? Where we're doing AI pipelines, where data needs to live in multiple places, but stay and think.

And so the beauty of that is that by separating metadata, we know those changes can happen instantaneously while we deal with the payload of those files moving to those geographic locations. So we can actually do, say a Ls on both sides of the world, see everything perfectly, but the data may still be in transit from a physical standpoint. Thank you.

So one other emerging problem that we've got in real time, um, is dealing with flash memory, right? And there's currently a pretty tight market for flash memory, whereas maybe a year ago, storage vendors were out there saying, oh, disc is dead. Well, now that nobody can actually buy flash or flashes going up to X in price because of this really tight market, um, we've got issues, right?

And so if you can't even acquire new flash, but you still need high performance storage, what do you do? Right? Um, and you know, there's all sorts of, um, ways that the system vendors are going about this.

You know, let's go out and do a buyback program so we can get some more flash in and then resell that to other customers. Um, you know, I I think it's a pretty, I don't wanna make light of the situation 'cause it's a pretty dire situation for, um, customers who really, really need that capacity. Um, but you know, our view is that there is actually underutilized flash storage probably out there in many environments, right?

You've stood up these multiple systems, they're all disconnected, they have an application running on system one. You have application B running on system two, application C, and all of those consumption patterns are not the same. So if you can aggregate those systems and use orchestration to make sure the right data is sitting on the right tier of storage, then you can free up some flash space within your environment.

You can make sure you're utilizing all of that flash space that it's already sitting there. Um, within the systems you have, we also have the capability, and we talked about it at the last, um, field day presentation, uh, a relatively new one that's, uh, was introduced about eight months ago, and that's tier zero. So that's the ability to consume the flash that's within the compute cluster, CPU and GPU co clusters and aggregate that within our namespace.

And so we set that up as a tier zero. It's a high extra high performance tier because you eliminate all of the network latency. And that's sit, that's the flash that's sitting in the GPU and CPU servers.

And we can incorporate that as part of the namespace and use, utilize that as a tier of very, very high performance storage. And if you look at, um, most of the GPU and CPU clusters out there, they do have underutilized local flash storage that we can access and add to the global data file system. And then finally, I mentioned this already, um, but treat cloud as an extension of the infrastructure, not a destination.

So, right, the cloud vendors have pretty good allocations of flash storage, but they aren't running into just availability problems. Now, probably not cheap, but at least you have access to that cloud instance, and you can treat that as an extension of your namespace that's both on-prem and in the cloud. So how do we do it?

So you can deploy hammer space, it's software defined. Um, we do have reference architectures and look for more news coming soon on other ways to deploy hammer space. Um, but we can deploy on virtually any server in front of any storage, including local NVME as well as anywhere, right?

Within an environment. So data in place assimilation here was Ray's question, right? So data stays in place and we assimilate the metadata into our anvil server.

I'll have a little diagram of that here soon. And data is visible to users within minutes, right? So again, we're not doing an onerous long-term migration pro process.

We're assimilating that data in place and then users see everything that they're authorized to use. And from there we provide a parallel file system access to a cluster or standard S3 N-F-S-S-M-B access, um, to other clients that are, you know, heterogeneous within the environment. And so we can accelerate data that has been sitting on legacy file systems or object storage that's not tuned for high performance workloads that the GPUs demand, while also managing all of that storage underneath.

And then finally, right? I think the most important point here, and the thing that we are going to exploit when we talk about, um, the A IDP integration with Nvidia is all of the automated orchestration that we can do underneath the scenes, right? So as Sam mentioned, we can move data right across locations, between systems, even doing so while it's being written to, and, um, we can set up policies based on, you know, normal business language that allow people to tier their storage as necessary based on their individual requirements.

Hey, Kurt, it's Ray Lui, Silverton Consulting, you mentioned earlier MCP. Do you provide, um, MCP server for agents to be able to move the data based on where they think they might need it, things of that nature? We Do.

And so Sam, that's exactly what Sam's gonna be talking about with our A IDP, um, solution. So that's where it really turns into something truly, I think, mind blowing, where you go from this really, you know, complicated management schema and, you know, none of this is easy, right? I don't wanna make it sound like, you know, under the covers everything is super seamless, right?

But we go from something that, you know, takes some storage management and storage intuition and experience and are able to turn it into va basically a, a system that can interact with you via natural language. So we'll talk about that a little bit. So, um, here are the building blocks, right, of the hammer space deployment.

So you've got the metadata control plane with the anvil servers, and then essentially the data service nodes. And those are the things that provide file access as well as do the data movement. And so we can use PNFS, right?

2, which is integrated into every current Linux version that's out there today for folks to do high performance parallel file system access to the data within a hammer space namespace as well, again, as your standard multi-protocol file and object access for folks who are doing heterogeneous clients, you know, for all sorts of, A little bit about how you've clustered these various different types of nodes as well. I mean, I, I assume that you have some way of clustering them so they, there's redundancy and, um, you know, Yeah, Sam, you wanna The more availability Yeah, exactly. There, there's h chain durability, right?

Built into the architecture of these platforms. And so part of this thing kind of like a, a ring architecture, right? And so part of those anvils will live in each site, right?

Controlling local metadata and the storage that's there and those, you know, and then replicated it in other areas in those particular sites, right? And so that's kind of how Kerr talked about how we do this out of band, right? But yeah, we, we cover all sorts of reliability and durability, right?

999 however far we get in that particular architecture, right? Like most systems are always a trade off between, you know, how transients that data, how much do we care, where do we want to do, especially with cloud resources, but our architecture is fully redundant a across the entire platform, I guess, you know, follow on to Andy's question. Um, so the redundancy, so some storage systems have redundancy built in, some do not.

Um, you're talking, I assume about the anvil metadata being redundant. I understand how that would be completely under your control. But let's say I have some, some dumb server storage and I want to have some sort of, uh, high availability for that.

Do you provide, uh, uh, I guess I'd call it external rate across? Yeah. So we do eraser coating across those stor.

We can do it across storage systems, right? To provide redundancy Okay. For the data.

Yeah. The, the other point real quick is that all this is done via policy and orchestration. So by share, right?

So not even by storage box, right? But by share by data type, we can choose to keep multiple copies of that data spread across different geographic locations or to Kurt's point, whether it's a ratio coded or whether it's stored locally across a handful of servers. So we've got a lot of flexibility in how we do that very granular, right?

So it doesn't even have to be a master policy. We can do that specific to projects specific to shares or specific to organizations. Yeah.

I my question was more specific, specifically how you handled the cluster of your, your metadata nodes and you know, the, I I think that you answered it adequately that, um, you know, there's, they're fully resilient. You, you, um, if you lose any one of them, you still find, Correct. Yeah.

They're, today, they're in an H eight care. We're changing that to be fully containerized and scale out. So we'll be able to add performance as we add containers.

Um, same with the DSX, right? Those data movers and some of those components, right? We can scale those today.

In fact, we already have some customers that do an early access in that containerized format. But today it's ha But to your point, that's fully replicated across our environment. Marian Newsom, I have a question.

How do you prevent, uh, approved access on those different sets of data? I'm sorry, can you say that again? How do you prevent, how do you keep the privilege access the same across those different sets of data?

Like the S3 Or No, exactly. So, so this is kind of the beauty of a single namespace, right? Is that right now those are all managed silos, right?

We see people sometimes do direct permissions into an individual na active directory, those sorts of things. By actually tying those directory systems into our platform, we a inherit those default permissions, right? So that's part of understanding the POS landscape perfectly.

So we can pull those particular things in. But, and I'll talk about this in my section a little bit, but the beauty of what we're, we're aiming for, right? Is that security, is it tied now to where data is born that that security can now follow that data?

And that's the beauty of separating the metadata from the storage itself is that now those permissions can follow that data whether they live in the cloud, live on a different storage platform, et cetera. And so we move that security plane into our metadata, global namespace realm. And so if I'm a, a company with cross-border data, uh, concerns or AI regulations, uh, that's what that solution will provide Exactly.

We can set up all sorts of policies, whether it's GDPR to prevent things across country, whether it's certain patent information, we can do some custom metadata tagging. In fact, we're actually doing some integration with some, some digital security posture management companies and stuff as well, right? The Tek, even our file system attributes well beyond that to be able to really control at a granular level what data should or should not live in certain places From ever audit it.

Could I ever see the logs to that? I'm sorry If I get audits, can I pull the logs from my regulators? Correct.

Yeah. The, the whole thing is fully audited, right? Where files have moved, who's access to everything.

That entire premise is built into our solution. Thank you. And I think there was an important point, um, that, you know, uh, Sam made there that maybe have been lost and I hadn't mentioned yet.

And that is, we definitely go well beyond a storage system in terms of the amount and the type of metadata that you can attach, um, to any object within our, um, global namespace. So that allows you to take additional steps, right, based on business rules to be able to interact with that metadata. So it's not just, you know, your posics metadata of file create and permissions and things like that, but you can attach other attributes onto that data, which again, becomes important when you get to managing a system for ai.

Um, I'm gonna go through this quickly here, um, just so we can jump over to you, um, the next section. But, um, I wanna talk a little bit about how our customers are using Hammer space today. Um, here's an example of a large digital payments company.

Um, and they had a team of about 3000 data scientists and research engineers who were building AI applications, and they were looking to move off of their fairly expensive up for renewal n systems, and they wanted to deploy object storage, cheap and deep storage, you know, cheap and wide storage, I guess, um, for their data scientists. And they started this process and then figured out that, oh, our data scientists don't speak S3, they speak NFS, you know, so what do we do here? Right?

And, you know, there's systems that you can put a, um, bridge in front of an object storage, you know, those are integrated to more or less extent, but typically don't provide you very good performance when you're accessing that object storage through some kind of gateway, but instead with hammer space, right? We can put that in front of the object storage, still take advantage of that scale out architecture, but now provide parallel file system access or a cluster as well as all of the other, um, access protocols natively, right? And so we also extended then into GCP so that they had burstable capacity for additional GPUs, um, when and if they need them, um, as well as now a fully tiered environment that was NVME or tier one storage, and then this object storage on the backend and cloud as needed, reducing storage costs, their renewal license basically, um, came down by $5 million.

Um, you know, the other thing is now their users are only looking for the data in a single place, right? They aren't looking across file systems, across object storage, it's single global namespace, no matter where the data lives greatly simplifies their workflows. Um, and now they have agility to be able to utilize both on-prem and cloud resources.

Um, here's another customer who ran into the performance issue, right? They were looking to deploy this new set of NVIDIA servers that now demanded, you know, three, four x the performance that they had been providing from their storage systems. Um, but they weren't looking to, you know, up their storage by three to four x.

Um, so instead of buying a bunch more storage systems or perhaps, you know, some kind of bespoke parallel file system solution, right? They brought in hammer space, we could still continue to leverage the systems that they had invested in their NAS systems. So those stay in place and be used both with data in place and then also as an extra archived here.

And then we just deployed our own NVME storage behind. So I did want to emphasize that we don't just use other people's storage as a, um, backend, we can deploy our own storage as well. So we do have, um, storage boxes to be able to support customers in their deployments.

And that's just standard commodity based storage servers. Yeah. Product Van Hern from, uh, ENS Consulting.

So, um, I, I do definitely see the value on the training sites, you know, with the metadata being available all over the place, but your data might be delayed. Uh, how does that work for inference? Do you see the product working well on, on inference considering that your data might be slightly delayed?

You know, where we talk about a little bit latency, where on the inference side, you know, latency is kind of a key, uh, key factor. Yeah, I think, um, I mean, I guess fundamentally, um, what are you changing from data being remote or local, you know, in the situation, right? We are gonna bring the data as close as possible to the GPU, right?

For, um, you know, again, under the covers can do that based on policy as a object is, you know, accessed. We can then bring the entire project into the local NVME. Um, so I don't think we're necessarily at a disadvantage.

I mean, yes, if you had migrated all of your data into a bespoke file system, right, and then connected that up to your GPUs, um, then yes, but I think then you're looking at a massive one-time migration, um, which takes time, right? So it depends on where you wanna spend your time, I guess. So If I understand it correctly, there's a way to create a profile where you kind of preload the data.

So Sam will talk about that. Okay. Here in Jeff.

Yeah, Frederick, we can spend some more time over this. We do have a number of customers. They're using us for global inference, right?

So when they're using tier zero for incredibly pa fast performance, right? Holding model repositories information that's been created, we then use a bunch of policies to drag that back, right? To continue to do some fine tuning or some updates of those things.

But they also use us to, to sync vector database information and those sorts of things. And at that point, we're up against the speed of light problem, right? But they still find value in the fact that this all happens through policy and automation rather than this like spray and RC problem they've been running against.

So I can show you a couple designs where we're helping those particular inference farms, especially when they're trying to, to your point, keep latency low by moving inference close to the user, but still keeping data sets intact right across those particular farms. Yeah. Thank you.

All right, I'm gonna skip over this slide. It's what we do. We're helping tame the data chaos with a single unified global namespace.

Now you'll notice here, right, I'm just talking about the unified namespace plugging into an AI factory, but there's an important part of the prep and rest of the data flow, um, or the pipeline, right? For AI data that maybe we hadn't touched in the past. And that is getting data truly AI ready, right?

We were a storage system that provided data, raw data at the front end. Now with A IDP, we're truly unlocking all of the value that is within the data.