Techstrong TV March 9, 2026

Hey, everyone. Welcome back here to Tech Drunk tv. I'm happy to introduce you to my next guest.

It's his first time on Tech Drunk tv, so let's welcome him. Warm warmly, it's Bob Bobo. Bob is the founder and CEO of Koft.

We're gonna hear all about it. If you're not familiar with Koft, it's okay. Hey, Bob, welcome to Tech Drunk tv.

It's great to have you on here. Alan, thank you very much for having me and, uh, appreciate, uh, the conversation we're gonna have. I well, I hope you say that when we're done.

Yeah. Good. I'm sure I'll, Hey, Bob, as I mentioned, you're the founder and CEO over here at Koft.

Mm-hmm. Give people a sta like, I've, I've founded a few companies in my day as well, and no one wakes up in the middle of the night and says, I want a founder company tomorrow. Right.

It's something, you gotta kind of feel it, you gotta feel it in your guts. You gotta feel it in your heart and your head. Right.

What was that feeling for you behind Koft Y? Yeah, great, great question. I was in a very lucky, uh, position in my career at several different companies where I was involved with, uh, building software products.

And part of that was not just building the products, but looking at alternatives to the products and in a competitive situation, but also looking for acquisition targets. And so, what's always interested me, and I come from an entrepreneurial family, uh, is how people got to where they are, the mistakes they made and the things that they did that made them successful. And I will tell you, uh, this isn't my first company either.

I've had three or four companies, and, um, you learn a little bit each time. Rarely, you know, are you, uh, a one hit wonder with these things. It's usually several companies, and then you, you mature the way you approach it.

And, uh, this time I think we did pretty good, pretty well. Very cool. You know, in prior companies prior experience, give us a sense of kinda where your expertise was doing, you know, what, what you came from there.

Well, I have a, a really short sort of history, uh, backstory here that I'll give you, um, because I think it, it is usually pretty interesting to folks. People don't often, you know, have a job that they started out or intended to have. I originally intended, uh, to be an aeronautical engineering.

I went to the Ohio State University. Um, had to give them a little pitch there. Absolutely.

And, uh, they have, They can have three of six players, the first six players in the draft. Three may be from there you go. Yeah, exactly.

Exactly. Uh, so, um, I did not finish my degree, but, uh, uh, in engineering for various reasons. But I did start my first co software company there.

It was a cad cam software, uh, product, and it ran on Windows 2 86, if you didn't know, there was never a Windows two. It was Windows 1 2 86, and then Windows three and like three. It was right as they were transitioning to three.

And so I knew a little bit about Microsoft and, and, you know, back when they were 75 people, not the juggernaut they are today. Mm-hmm. And through that, I swore I'd never do another software company.

So I started doing consulting and programming, uh, and networking very early on. This is when businesses were just starting. I'm in Columbus, Ohio, by the way, still.

Mm-hmm. Still even after college. Hey, big tech area board in Big Tech and all those folks.

Yep. And, uh, so I woke up one morning, my wife was sitting on the edge of the bed. She said, Hey, I love that you're your own boss and that you're doing consulting, but, uh, we're gonna have a kid now you need to go get a real job.

Hmm. So I went to work for a small startup called IIDA Software, and they were in the Windows tool space, and so it was a natural fit. Uh, several years after joining them, uh, they sold that company to a company called Quest Software.

Sure. And then for the next, uh, eight and a half, nine years, you know, I enjoyed working for a company that was both, uh, leading that space, doing acquisitions in that space. And, you know, for seven of those eight years, we were Microsoft Partner of the Year.

So, you know, worked with the largest and medium, and even some of the smallest, uh, Microsoft customers really learned a lot. Uh, they decided that they were going to sell to Dell and become Dell Software. And I thought, you know, I've been here for a while.

I wanna try something new. And so I went to another company for six or seven, eight months in the similar space. And at some point I was like, you know, Microsoft is announcing this big cloud shift.

You know, they're moving from on-premise client server to cloud. And my, my current business partner and I both recognize that this was the paradigm shift that we had seen all those other successful, uh, folks, uh, catch a hold of AIDA was there when Active Directory started. Um mm-hmm.

That was started by Mir, Tim and Andre Bernoff that went on later to create Veeam Software. Uh, and we really recognized this pretty quickly. We also recognized differently from a lot of other folks, uh, that this was not about enterprises and moving those to the cloud.

It was really about protecting the office, um, uh, product line from Google, the franchise Google Works or Docs and protecting the server market from Amazon. And so we realized that the enterprise customers that we had served for, you know, the past eight, nine years, were not gonna have the tooling that they needed to make this transition. And yet they all were being pushed to move their, their expenses from capital expenditures to operational expenditures and use these cloud things.

So, you know, for those of you in the audience that ever had to work with Microsoft Exchange, it's a very complex email product. Right. And being able to offload that onto that, that piece of your infrastructure onto the people that wrote the tool, is a godsend.

It makes it that much easier. Right. And, and even with identity, there is a lot more energy put into the security side of Entre id, which is the next generation of Microsoft's directory, uh, versus active directory, because it's on premise.

Microsoft doesn't control the security there. You're on your own to control it. So at some point in the future, you'll see that those other, other, uh, pieces shift for a large segment.

But right now we're enjoying this, this time where we recognize that enterprises are locked in this journey. They started with Client Server active directory and exchange on premise and SharePoint on premise. These client server designed and developed tools.

They have some really great collaboration and additional, um, um, operational benefits they get by integrating Cloud, but they may not be able to go fully cloud. So they're in this situation where they have a lot of on-premise and a lot of cloud, and they operate these two separate environments. They don't look the same, they don't act the same.

They have different security paradigms. Uh, and we bridge that. So we're actually a great way to take that journey from, uh, uh, on-premise only client server into hybrid mode, where we make it still appear as one cohesive environment.

And then eventually if you, you know, decide you're gonna go cloud only, you basically just turn off the client server components in our tool and you continue to manage. So it makes it a very, um, um, good, uh, uh, solution, uh, to solve some of the operational inefficiencies that hybrid, uh, naturally brings. Sure.

Um, and it, it's not just about management for us, we also recognized that the traditional model was to get into the market with a single tool like Active Directory Recovery, for example. That's one of the tools that launched IDA and got them, uh, acquired by Quest. But the old model was, Hey, you know, we have this market.

We're selling to the active directory guys and the infrastructure team. Sometimes it's identity, sometimes it's security, but, but we have that, that market, right? And it's easier to sell another tool to the same people.

So let's go build a second tool and then a third tool and give it to the salespeople. They call it putting it into the salesperson's bag, right? So that salesperson can then just go and start to sell more to the same people.

Well, pretty soon you have 9, 10, 11, 12 products. You have to build them each time. Developers build it a little bit differently.

And so they're not the same. They don't share a security infrastructure or a look and feel or approach. And at some point in really large companies, you might get away with that because the identity team now owns Active Directory, but the SharePoint pieces are over here.

And you know, now with Teams or Slack, there's another team for that. But when you get to the mid-size customers and smaller customers, it becomes very difficult. And so when we took this journey, my partner and I looked at what hybrid meant, moving those resources, you know, over 10, 15 years from on-premise only to cloud only with this hybrid mode in between.

And we realized that we didn't want to lose the benefits of a completely integrated platform. So we don't just do management, we do management change monitoring in a, in a novel way. And then we also do recovery, and we do recovery in both rolling back changes to a Microsoft team security policy or changes to a group membership that's all part of and integrated into our change auditing solution.

But we also do forest recovery. And that's where we really started to see, uh, adoption in the market of our solution and interest. Uh, prior to that, there were only two vendors in the space that could do a Forest recovery Quest software, where I was, where came from.

I understood that. Matter of fact, when I'm, my, uh, customers, uh, when I was early days at Quest was one of the first Forest Recovery customers that, that Ali had, and they had to do it all manually. So they went through all those 400 and some steps, I think it was to, uh, help that customer recover.

And then that became the Forest Recovery tool. Um, but we wanted to get into that space. We recognized that that was a, a growing area because of ransomware attacks targeting active directory.

I think Microsoft says 80% of ransomware attacks target active directory. That thing was never meant to connect to the internet. Right.

And yet with COVID and work at home, it, it just happened. Sure. And so as we started looking in that space, we realized that management delegated administration, least privilege, zero trust combined with seeing what's going on in these hybrid environments.

And then finally being able to recover when you make a mistake or when somebody gets in to recover from a ransomware attack, was the, the three big, uh, buckets of, of, um, we'll call 'em buckets of help that, that IT teams modern IT teams need help with. And so that's where we concentrate. And we got really lucky with that last area.

Um, we were so small, there were only three of us in the company when we started working on recovery. And, uh, my, my business partner called me and said, Hey, we're gonna announce forest recovery, this, this, you know, ransomware recovery thing. Um, I was in Florida at the time, I realized that you're down there in the warm weather, in the snowy Columbus weather.

Right now it's, it Is. I don't wanna rub it in, but it is warm today. Yeah.

Yeah. So I was in southern Florida at the time, and he called me late one night, and we had a demo. I think this was Saturday night, 3:00 AM he called me, um, he's in arm, was in Armenia at the time.

That's where our development center is. And he, um, he said, Hey, Bob, we got the forest recovery thing ready to go for the demo, but you're gonna need four more laptops to demonstrate it for the recovery piece. And I'm like, I'm in nowhere Florida.

There is no Best Buy within 300 miles of me. There's no way I'm gonna get to that by, you know, Wednesday or Thursday and get all that set up. So we had a, a quick discussion and realized that we had all the makings to do it all programmatically in the cloud.

So we, you know, mother's the inventor of necessity or whatever that saying is, um, mother of invention, Mother city is the mother of invention. That's It. That's it.

And so, um, because of our size and because of our limited resources, we had to do something different. And it turns out it was a game changer. Um, and so we're likely the fastest way to recover in those, and probably the one that you have a hundred percent certainty is gonna work.

Um, Fantastic. Yeah. It, it really worked out well for us.

And, and, you know, when we got that thing to market and started demonstrating it, that's when the acceleration really took off. I love it. You know, Bob, as I sit here and, and I'm listening to you, the two things that I missed in when, with the rise of cloud computing, the, well, the biggest thing I missed is I never saw multi-cloud coming.

Mm. Right. I, I, it just didn't occur to me that you would spread your infrastructure across multiple cloud containers.

It just seemed wasteful. I gotta learn three different platforms, what have you, but yet here we are. Right?

Right. Each of the major cloud providers have their own special sauce, their own things that they may do better than some others. And so, so many organizations today are multi-cloud.

They, they do use Microsoft Office. They do use Active Directory, or hosted Active Directory, and Toronto, I forget what they call it, ra. Yeah, yeah, yeah.

Mm-hmm. Others are using Google for what have you. And then of course, others, you know, they, they first moved to the cloud on Amazon and they still have instances there, and there are things AWS does that the others, you know, don't do as well.

So, you know, multi-cloud I think was, so the second thing was I thought the train would keep going. Like, I didn't think everybody would move everything to the cloud all at once. Mm-hmm.

But I thought once you set up shop in the cloud, there would be like an underground railroad, if you will, of things constantly, until finally you emptied out your, your server closet. Sure. Now, it hasn't really worked out that way, has it?

Right. We still have, there's probably, I'm gonna say still a majority of, of applications and infrastructure deployed not in the cloud or not in the public cloud, and may be at third party data centers or what have you. Right.

Colo, but not in the third party cloud. And that to me is, that's gotta be music to your ears, right. Because the longer, you know, the more we have in these hybrid environments, the more they need the kof type of solutions.

So there, there are two things that I think, um, support that, that, that thought you just had. One is that I think people don't quite understand the magnitude of Microsoft's previous generation of client server. And when I say that 90, and this is a Microsoft number, 96% of businesses worldwide ran active directory, that number, I no doubt it Is almost unconceivable or inconceivable to a lot of folks because they don't really understand how, but if you went all around the globe, everybody used Active Directory.

And so I think there's a scale that it's just gonna take time because they had built for 25 years, by the way, happy birthday to Active Directory, which was two weeks ago for 25 years, people built applications around active directory. It was the, the, the standard. No, it's a monopoly.

So it takes time to unlined all that. Right. So there's, I don't want to use the M word.

We'll get sued. Yeah. But it, it, it pretty much, it was the standard.

I mean, I'm old. Look, I remember a time of NetWare. Sure, sure, sure.

Or Banyan Vines before that. Yeah, yeah, yeah. You and I with the same league here.

Yeah. You know, so the LDAP thing, I think it just is taking time, you know? Mm-hmm.

So I think that's an important one. And interestingly enough, when we started this company, uh, and we started to discuss what was available, uh, to us in terms of what we thought we would do for product, we were gonna start with Google because Microsoft had not yet announced Office 365. And so the platform we built actually is multi-cloud capable for management, monitoring, and recovery.

We have gone down the path very similar to what Veeam had done, chasing VMware first and then switching over to Microsoft and doing both. We are in that process currently with our platform of embracing what we've known the best, which is the Microsoft stack, but our first betas and testing role on the Google stack, interestingly enough. So I think Any, any plans, and don't say anything you're not supposed to.

Sure. But any plans to, to work with the Google Docs? The Google office?

So we already do, we already do. Oh, do we have several customers that we support? Um, uh, unified user provisioning for and group management.

We also support, uh, AWS for either running the solutions that, that we have on the client server side. Mm-hmm. Uh, or, uh, for example, with Forest Recovery, you can actually use an AWS instance to further insulate you from your on-premise active directory.

So you can have our automated backup and recovery of your directory. We actually call it a standby forest. We clone your forest every night.

You can do that in, um, uh, Microsoft's Azure, which is probably what the majority of folks are doing, right? Mm-hmm. So you have that clone of your ad infrastructure there ready to go, should you get attacked, or you can do it in Amazon.

Um, and we also support some on-premise, uh, as well. But, uh, yeah, I think as, as we see customers coming to us, that is how we know when we need to start supporting things. We don't just, you know, put our finger in the air and say, oh, I think we should go do something.

We're constantly taking feedback. And I think that is another differentiator that's allowed us to, to really accelerate our growth. You know, a lot of companies say, Hey, we're doing great.

Um, we did about 76, 70 7% growth in our revenue last year, coming off another year. Uh, and we will likely continue that or maybe accelerate that a little bit more this year. And I think that combined with our, you know, high 90%, um, customer retention proves that we're on the right track with our unified platform and listening to customers.

And as they run into these challenges, we're hearing that, and we then embrace those things. And if they come back and they tell us, Hey, you need to support Jump Cloud or Okta, that'll be the next thing we do. Right?

So for me, those things are, are going to be revealed to us through our customer base. And our customers are, you know, mid-size Microsoft customers at the moment. Right.

That, that's what I was gonna say too, there. And, and look, in a lot of ways, I, you know, it was kind of a, a, an unspoken truth in enterprise. Microsoft controlled the enterprise, right?

But they also controlled the mid-market and small market. 'cause they just made it so easy. E look, even, uh, when I started, uh, tech UNK 12 years ago, 13 years ago, I mean, Google was there, don't get me wrong, but we were micro, it was so easy to just use office already.

Right, right, right. And it was a no brainer. Right.

And we, I didn't have on-prem, everything was sas. I didn't have a server cars. I, I still, we have a room here in the office that still has, when we moved in, it had a, a sign on the door, it says Data room.

I got a cardboard cut out of Commander Data and put it in there, and that's data's room. But other than that, that's, that I got nothing. That's funny.

Uh, so, so I get it. However, I will say that, look, we're ad, I don't want to throw the M word around, but you're right. 96% of the world, I think, was on active directory.

Um, there are other directory choices. I think Google and I forget what they call Google Office, the official word for it, but Workspace used to be Docs. Yeah.

Workspace, Google Workspace Soup, dejour is this week, I think. Yeah. You know, it, it's made tremendous gains and, and inroads.

And, you know, we use it here too. I, I offer people pick whichever one you want. I'm not, you know, discriminating, but, um, exchange Server as well, right.

Versus Gmail, corporate Gmail. It, this is the world we live in today. Right, right.

And, and I think that's why you need companies like KioSoft to, to navigate those borders, to navigate, you know, whether I'm gonna be all on-prem, all in the cloud, a little of both, a little here, a little there. We, we need something that stands out, those rough edges. And, and that sounds like, Yeah, security identity infrastructure teams face a, a, a major upheaval moving to cloud, and, and it's not a bad upheaval.

I think, you know, with Google coming to the forefront, it challenged Microsoft. Their stuff has gotten better, you know, because without a challenger sometimes, you know, you, you languish a little bit. And with AI coming on, they're embracing it.

You know, You have to either you well really embrace it, or you're gonna get rolled over that, that can't believe it's not there. You know, you're talking about embracing a challenge. I remember when Microsoft invested in Apple.

Right. Remember that? That was crazy.

It was good. Yeah. But they, well, they had to have a, a, otherwise the Justice Department might have got involved, right?

Yeah. It just took justice a little longer than I think after, well, Whatever. Look, they, those guys at Apple did okay for themselves.

Yeah. com, CAY off My t-shirt right there. CAY.

Yep. And well, it's also gonna be in the bottom third that we have under you. So it's all in there.

Not to worry. Um, for people out there say, yeah, I'd like to give this a shot. We, we are in this kind of, that's exactly describing our situation.

What's your, excuse me. What's your best advice for them to get started with Koft, Bob? So, you know, I, I think if you look at what your Microsoft infrastructure is delivering, if it's around identity, if ongoing management or recovery, we're probably the, the, the, the number one choice right now.

Um, you know, we do a lot of, of replacement business for other companies where you might have a legacy solution. So, you know, come visit us there if you wanna just kind of see how the products work. Uh, we actually have a freemium tool called Guardian Protector.

And Guardian Protector allows you to run security scans a lot like antivirus for your identity. Think of it that way. It works with both, uh, active Directory and entra.

That's a great way to try us out for free, keep using it. We love it. We update it almost weekly with new threats like an antivirus would.

Um, if you're, you know, really struggling with like user provisioning in, in ad in a hybrid environment, you know, uh, contact us. We're happy to run a proof of concept with you, or, you know, if you just want a demonstration, we can do those things as well. com.

I love it. Hey Bob, thanks for coming on Text on TV today was awesome. Thank you so much.

Hope you enjoyed this. It was awesome. Here as well.

Come back and keep us posted. Best of luck with Koft. We'll be watching.

Okay. Enjoy that warm weather in Florida. Well, it was, it's, we've had cold weather.

This is the first week it's actually warming up again, I hope. Awesome. We're done with the cold weather.

The one is, we're falling outta the trees. Alright, thanks Bob. We'll be in touch.

Thank you. Cheers. Mm-hmm.

Bye-bye. We're Gonna take a break here on Text Trunk tv. We'll be right back.

Hello and welcome to the latest edition of Digital CXO Leadership Insight series. I'm your host, Mike Bazaar. Today we're with Israel Forced, who's the newly appointed CEO for valis, and we're talking about the state of digital transformation, which I think we've been at now for at least a better part of a decade.

Israel, welcome the show. Thank you, Mike. Thank you for having me.

It's a pleasure to rep, uh, VALIS here. So, uh, you've been with the company for a while and you're kind of the newly appointed CEO, but what are you seeing out there? What's kind of the goals for the company and where are the gaps that you guys are gonna plug?

Yeah, no, it's a great question. So, you know, VALIS has been around for about two decades, and we've spent the last two decades really helping companies, uh, and organizations modernize how teams collaborate, how they deliver work. We've done that by implementing and scaling, you know, leading platforms to support how work gets done.

Um, and but over the last year, I would say we've seen, we've seen a clear shift in our customer's priorities, CTOs and heads of engineerings. They're not just asking about platform modernization, they're not trying to get to the cloud or get in the latest tool. Uh, in fact, I would say the reverse, uh, the opposite is happening.

They're being inundated by just a plethora of new tools, um, new capabilities, new vendors out there, uh, and a lot of 'em are getting stuck and trying to figure out how to navigate this rapid evolution of the developer experience. Uh, the market's crowded, there's a lot of competing narratives, there's a lot of hype, but I, I think we'd all agree, there are very few companies that have actually achieved tangible enterprise grade results, uh, in this sort of, uh, in this new world. So our focus is shifting from being a platform implementation company to really focusing on execution performance in AI accelerated world.

Like how do we help organizations redesign how workflows through the very functions, uh, of the organization in this new and exciting world. Well, to your point about that, how much of these challenges are really technology related, or are they just as much cultural and kind of maybe even old fashioned inertia? Yeah, no, I, I think, uh, you know, it's funny, I was talking to a colleague of mine and we were, you know, it's often, I, I, I think a lot of us are students of history look back and, you know, they say history doesn't repeat itself, but it certainly rhymes.

So you can look back at a lot of these, um, big waves of change renovation that have come through and ask yourself like, what prevented adoption and value, and then what ultimately unlocked adoption and value. And, you know, we can go back in this context. You can go back to, uh, as the org as, uh, organization shifted from Agile to, from waterfall to Agile, and you know, how that started and what, what did it take to finally hit this tipping point where we started getting real acceleration from it.

Um, you can go back to the, uh, industrial, you know, engineering and how they've modernized factories. Um, in fact, I was listening to a podcast the other day with, um, uh, Jim Farley, the CEO of Ford, and he was talking about what he needed to do to really transform Ford. So there's a lot of, I mean, this idea that there's innovation and rapid changes that are, uh, capable, and yet it doesn't seem to change the world in the way that we want it to until something happens.

And you ask yourself, what is that something that happens? And, uh, I think if you look at a lot of those examples, and you can probably apply the same thing to AI, is there's a lot of optimization in context. You know, you have a team that has a set of processes and you say there's a new tool that's wildly better.

And so they look within the way they've orchestrated things. You, you know, you can apply it to the, the world of agile. If you simply took one team and said, be more agile in a waterfall world, it, it, it, it would be, you know, 1% better, 2% better.

If you went to a factor and you try to automate one step or one person's process, you would get incremental improvements. But it's not until we're ready to rethink the whole process from end to end and start with a clean slate, that I think you get the real benefit. And you know that, I think this is why you're seeing, uh, uh, a, a divergence, uh, between some of the, what we call now called AI native companies versus the companies that have been around for a while because the AI native companies, they have no baggage.

They can just start with all the capabil bills that exist today. Whereas the companies that have, you know, years and years of legacy capabilities, legacy ways of working, they have a much harder time, uh, changing how they fundamentally get work done. And therefore, um, they're, they experiment, but they experiment in compartmentalized ways.

And I think that to me, that's fundamentally, uh, why we're seeing this, uh, limited value in an enterprise scale and what ultimately has to happen for us to help customers get value at an enterprise scale. Everybody and his brother is talking about AI these days, and they're running into these issues. And some folks say, you know, it's the greatest thing since sliced bread.

And others are basically saying, I'm running into the same bottlenecks twice as fast. Alright, from your perspective, is AI at the very least gonna force us to revisit these digital transformation conversations that we seem to be, have had, maybe off and on, but not consistently enough? I think, um, I think it, so if you were just to compare AI to the its predecessors in terms of digital transformation, I think the, the, the multiplier effect that it could have is greater than others.

And I think that is certainly true. So if done right, it has the potential of dramatically compressing cycle times of, uh, dr I don't wanna say infinitely, but dramatically increasing capacity. And a lot of the constraints that you previously had, they're no longer constraints.

So I do think it's in some ways is different than some of the prior, um, uh, impetuses for digital transformation. But I also think inherent in that is it is going to be harder for us to achieve those, uh, benefits because I think it is gonna require that companies rethink completely how they do things in order to get that value. So I, in, you know, in some ways I think it's, it's bigger than its predecessors, but it's because of that, it's gonna be much harder.

And I think the companies that do it well are what, at least what we're seeing and what we're advising customers is they are figuring out how to, um, carve out end-to-end parts of the business. It's a particular product or a particular segment, and, and it's something that is, uh, high, high potential, but maybe low risk, but something that they could get an end-to-end benefit from and say, go rethink this in the world of ai, but do it in a, call it a pilot, but do it in a standalone, NewCo Spinco, whatever you wanna call it, so that they can see what would it look like without the baggage that we have. I, I mentioned Jim Farley before, but you know, when he was trying to get four electrified, that's what he did.

He said, let's take this part of the business. Let's, let's free it from the baggage of Ford and see what you can do if you were just your own company. And then how do you then impart some of that experience and learning back into the larger company?

And I think that's what it's gonna take, um, for executives to, uh, to drive AI benefit across the, uh, across the entire organization. And again, just to pull on that thread, it, it wasn't, you know, it wasn't some, uh, VP of manufacturing that made that commitment was the CEO of Ford that personally went and did that. And so I do think that's one of the other hallmarks of companies are gonna be successful are those where you have leaders at the very, very top who are not just mandating the change, but actually leading the pack.

And I don't say hands-on 'cause Jim wasn't actually manufacturing, but really driving or creating this space, if you will, for that change to happen within the organization. I think the same is gonna be true for, um, AI in the engineering and product development context. I think one of the challenges that I hear from people is a lot of the folks who suffered from the initial wave of a fear of missing out investing heavily, and then they woke up one morning and discovered that some application that they normally use in the app built in all the capabilities that they just spent the last nine months experimenting with.

And, you know, suddenly a million dollars went out the door, but then it turned out to be, it's a feature of something else. So do we have to get smart about what projects we're gonna actually invest in? Yeah, I do think, and that's, you know, that's why we spend a lot of our time is really connecting strategy to work and helping customers do that.

And I know we talk a lot about, you know, observability as an example in, in the AI world, and we think about a lot in the context of the technology. Like how do you observe what the agents is doing and how do you have certainty in a non-deterministic world? But I think it actually, it, you can look, I mean, certainly there's this observability in technical context, but there's also observability in the work context because you can scale bad decisions just as quickly now.

And so this idea that you have, like, you make a plan, you roll it out, you see what happens, you come back in a few months and observe it in this world, uh, you actually have to be observing that almost on a daily basis. So this ability to connect strategy to work, um, really has to happen much better and much faster because you will cycle off, you know, off strategy if you will, very quickly with the scale and capabilities of these new tools. So I think that's gonna be another interesting challenge for organizations is how do you, um, uh, move from a leadership perspective and how do you, uh, drive the, uh, company direction all the way down to the person or agent that's developing the code in doing that in a way where you're, um, you're actually keeping everyone flying in the right direction.

And I do think though, um, the, the counterpoint to that is it does lower the cost of getting it wrong because, you know, you can, you can, because, you know, capacity is not as limited as, well, I don't wanna say it's free, but it's not as limited as it was. It's not that hard for you to, as an organization to try five or six different things and get far enough to be validated. Whereas, you know, in the past that was probably a, a 3, 4, 5 month endeavor before you realize that we kind of made a mistake here.

So it is a double-edged sword, I think, as it always is with, uh, with transformation, it can, it, it hurts, but if you use it right, it can be in, uh, incredibly valuable. How do we navigate the, who moved my cheese conversation? 'cause a lot of the stuff around AI or even digital transformation in that matter, you know, always had this sense of, well, there's gonna be change in how does it affect my job?

And in the age of ai now it's become, you know, oh, well, we're gonna reduce our headcount by X, Y, z and you know, a lot of people kind of look at that with, uh, shall we say, less than enthusiastic response. So how do we kinda, you know, get people to kinda wrap their heads around how do use AI in a way that maybe benefits everyone? Yeah.

Well, I, I think, you know, I'll go on a little bit of a soapbox, but I don't think some of the, the narratives that some of the public company CEOs are taking out there, which is I reduce my head count X because of ai. I think that's, I think that's posturing a lot of their parts. Um, I don't think it's actually true.

I think they feel compelled to demonstrate benefit and value because of either investments they've made or because of pressure they're under. So they feel pressure to demonstrate that they're achieving value. But I don't actually think that's true or helpful.

Um, I I, and I think if you look, again, if you just look back at history, um, every other innovation, you know, step function innovation has not replaced, necessarily replaced people, but changed the, the, the, the layer at which people are being impactful and therefore allowed people to be far more scalable. And so I think, uh, you know, you can look, I don't know if you remember watching the movie, uh, hidden Figures, um, about the nasa, um, you know, the Apollo missions, and I forget the name of the, the woman who was the, um, she was the head of the computer department, which were the people that did computing. And um, there's a scene where, um, she comes in and she tells her team, they just put an IBM mainframe in here that can do 27,000 calculations in a second.

And the her team said, oh my God, we're gonna be out of jobs. And she's like, no, we're gonna become operators of the machine. And I think that mm-hmm.

You know, again, it was, you know, 50 years ago now. But I think the same is true. The people that are going to learn how to leverage and harness those tool and become more harness engineers as opposed to developers are gonna be able to produce, I don't wanna say infinitely more, but substantially more in the same amount of time, and the people that are gonna struggle to make that leap.

I, you know, unfortunately, I think there are gonna be some people left behind that always is gonna be the case. Um, but I don't, I don't think fundamentally it means that we're, uh, uh, eliminating positions. I think it's just a, a matter of changing where people spend their time and learning to become multipliers, uh, for the technology.

So what do you see digital CXOs doing time and again, that just makes you shake your head a little bit and go, folks, we might wanna be a little bit smarter than that. Well, you know, I think the, um, uh, a couple things. First of all, uh, trying to make vendor decisions right now I don't think is smart.

I think that world is gonna, I mean, you know, you, you think about, imagine it wasn't that long ago when we thought open AI open AI had this insurmountable headstart on everybody, and a year later the landscape changed completely, Gemini sprinted ahead, and now anthropic sprinting ahead. So I think trying to, to pick the winners on the model space is probably not productive. It's probably not where you spend your time.

Um, so I think that's one thing not to do is worry too much about that. I also think, um, while we want experimentation, I don't think experimentation is really gonna help us figure out how to get value from ai. I really do think it is, um, uh, finding ways to rethink your business and the bigger your business is, the harder that it's gonna be to do, that's going to be the, the challenge of these larger co corporations is how do I reinvent myself while still running my business?

And I think that's, to me, the idea of more experiments. Let's get everyone to consume ai, let's, let's measure adoption without actually defining what we're gonna change and how we're gonna change and why it's gonna be better, I think is creating a lot of confusion. I mean, you get a little bit of value, but you don't get a tremendous amount of value.

I think the smart C-suite individuals, uh, are leading from the, from the top. They're finding ways to carve out parts of the business and asking the, the, uh, the business leaders of those teams to rethink everything in this new world. And I think those leaders will make more progress than the ones that frankly don't have the courage to do it.

And I think that's, it's, it's, uh, it's gonna take a lot of courage for, uh, for leaders, particularly if leaders are large companies to, uh, reinvent themselves in this new world, uh, while managing the, you know, the legacy that they have. Right. So given all that, what's your best advice to those leaders?

Um, you know, should they walk the floor and get more deeper into the organization? Or is there something that they should be doing that they're not doing? Yeah, I mean, I do think it starts at the top.

I think if you look at, uh, I don't wanna say every, but certainly there are a lot of examples of real corporate transformations where real change was driven. And almost always it was because there's a, there's a leader at the top hands-on driving that change. You know, I spent a number of years at Salesforce and I, I, uh, remember the story of Salesforce moving from Waterfall to Agile, which was a, a massive change for that company back in oh 5, 0 0 6 or seven, I forget what year it was.

And it happened because Parker Harris, the, the, you know, founder and CTO led that change and, and not only did lead the change they came to and said, let's pilot it with one team. And, and he said, no, we are rolling this out through the whole company. We're doing it.

It's gonna be hard. It's gonna take months and months, and we're probably gonna make mistakes, but in the end, it's the right thing to do. Now, that was a huge bet.

It obviously worked well for Salesforce and for Parker. Um, but I think you think about the courage and the conviction that that took on his part, uh, I think modeling that behavior, maybe not quite so bold, but that type of behavior, I think is the, that's what you wanna model. You wanna model, uh, leaders who have a vision, they're convicted about it, they're willing to take on risks, they're willing to make mistakes.

Um, they find a way to create this space for their teams and give their teams permission to make some mistakes, experiment, challenge the status quo, free themselves of the baggage and the constraints that exist in the other parts of the organization, and then take those learnings and then start playing them back into the larger organizations. I think those, those are the, those will be the winners. I think the smaller the organization is, the easier that is to do, the larger the organization is, the harder that is to do.

Hey, folks, you heard it here. It's one thing to create the plan, but you know, it's another thing to do the work and there's no substitute for it. Israel, thanks for being on the show.

Yeah, thank you, Mike. Pleasure talking to you. Bye-bye.

And thank you all for watching the latest episode of the digital CXO Leadership Inside 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 time. Welcome to Reality Check with Dave Nicholson. I'm Dave Nicholson.

And, uh, this is a show where we do something that is, uh, shouldn't be as novel as it is. I guess. We actually talk to real decision makers who are spending real money, uh, real reputation points on making decisions in the AI space.

And, uh, one really important, often overlooked part of ai, especially among tech nerds like myself, who love to talk about servers, networking, storage, memory models, et cetera, et cetera. Um, is the, the legal area, uh, what are the ramifications that come from deploying AI in a large enterprise? Uh, what are the legal considerations?

We can sort of imagine some of those things, uh, but a lot of us, frankly, that are gonna be engaged with this, uh, think of the legal department as the department of, no, can we do this? Legal department says no. So I thought it would be great to bring in a very well respected, uh, voice in the world of AI and the law, uh, and the intersection thereof.

His name is Bill Price. Bill, welcome to the show. I'd like you to kind of give us an idea of, uh, your perspective and your background before we dive into this.

You're not just a lawyer, right? No. And, and thank you very much, Dave for the invite.

Honored to be here. And, um, no, I'm definitely not a lawyer. I've got 20 plus years experience in technology for both private and public companies, from startups to Fortune 25.

And if I was hired just to be a lawyer, I would not have been, uh, as successful as I've been fortunate enough to be nothing wrong, just to be clear, nothing wrong, nothing wrong about being hired just to be a lawyer. We, we, we, we love, uh, we love our lawyers, but, uh, continue, not at all. I'm proud to be a lawyer, but if you're gonna work in a business as I have for most of my career, you have to be a business person first, yes, for the legal acumen.

But what you're really paid to do as a GC is to get things done. And yes, of course, keep the company out of regulatory and other trouble, but you're really paid to help the other executives in the company exceed its objectives and help grow the top line, ultimately profitable growth. Okay.

So I must say, because, uh, I consider it a party foul when a TLA, whether that's a two letter acronym or a three letter acronym, is, is thrown out and not defined. So I will define gc, it's general counsel for the non lawyerly folks there. Um, uh, but continue, bill, what, uh, aside, aside from that, so, so from that kind of business acumen perspective, what are the, what are the, what are the areas of business that you have experience with?

So I worked for the first, uh, speed of ACU acronym, SaaS company software as a service before that acumen even existed. And it was to take a concept that an entrepreneurial, uh, founder had and make it real. And so we were starting from scratch, basically, but the concept was to take software and turn it into a service in the, what's now known as the cloud, which was really connecting to an offsite data center.

And so that was in the early two thousands. And back then there was no regulatory regimes or protocols to do that. And so we were going forward, um, with best practices internally, but realizing there was a lot of ambiguity.

And so what I got very comfortable with early on was working through that ambiguity and realizing at some point there may be a faux pa that you would have to course correct on, but that was what intrigued me and why I wanted to work in a business as opposed to, say, a law firm where I was working for one company as opposed to hundreds of clients. And I was responsible along with the other executives to exceeding objectives for the investors. And that's what really, um, attracted me to working as a general counsel, which as you said earlier, Dave is really the chief legal officer, the, the head of the legal department for a company, it's many more things than that, especially if you work for a startup, you can be the chief bottle washer, um, and you basically do whatever's asked, uh, by the company to get done and do it quickly and as you know, correctly as possible.

Okay. So let's, let's, let's jump right to present day. Uh, first of all, that whole SaaS cloud idea, clearly, uh, clearly it was never gonna work.

And you know, it, it that, that, that idea never got off the ground, did it? Uh, that's sarcasm from me. Of course it did.

And you were there, you were there in the early days. But, but here we are, uh, at the dawn of the age of, of AI terror within, uh, within large enterprises. What are some of the things that catch the attention of the general counsel?

Let's, let's pretend like we're talking about, you know, a decent sized enterprise, um, where, uh, the general counsel is not necessarily engaged in every single contract that's written, you know, in the firm. They're not the only attorney, uh, in the firm, let's say. So it's a, you know, it's a legal organization, but what rises to the level of the legal organization's attention when it comes to ai?

What are some key things that are concerns? Well, AI is just the latest technology, but as you know, um, anytime a new technology comes along, what the legal department is trying to do is much like any other department, three things primarily. There's not enough time, there's never enough resources.

And then the third element, which might be a little different than other functions, is the legal department wants to be known as, as you said earlier, not the department of No. And it can tend to have that stereotype. Um, so it rather would like to be known as the profit center, getting things done for the company profitably, quickly, correctly.

And so what the general counsel wants to do for the legal department is help move its culture, its reputation up the strategic stack. And so when AI came along, what in-house council were trying to do is leverage this great new technology to reduce the administrative burden, the grunt work, if you will, and automate that, have the machine do that. So it would free up the in-house counsel's time to be more strategic and work on business development as an example.

And that'ss really the benefit. Oh, okay. So interesting.

So, um, you're, you're, you're looking at it from the, um, from the perspective that, uh, the legal organization is going to leverage these tools themselves. Uh, what about the introduction of these tools into other lines of business where, um, where governance and compliance and things like that might be a concern? I guess it's, it's sort of an interesting situation where, where, um, you know, the general counsel and team may be using AI to police the use of ai, uh, in an organization.

What, what are, what are some of the things that, that you are seeing as concerns when a vendor comes into a company and says, Hey, we want to implement, we, we, we wanna enhance everything you're doing with artificial intelligence. And the line of business owner says, okay, well we need to have our legal team involved with this. What are the things that the legal team looks at?

Uh, is there anything different or specific about considerations with AI coming into the mix? I've got a couple of, I've got a couple of ideas in mind, but, but what do you, uh, what do you, what do you think about that? Absolutely.

It's like any, uh, third party solution really, the first question should be what problem are you trying to solve? And given the fact that budgets are tight, especially legal budgets, what we did that was pretty effective strategically was to look at the business case or ai, and if it was legal ai, how that could help other functions. And so it depends on the size of the company, it depends on the budget, it depends upon the culture really.

But from a legal perspective, the in-house counsel has to realize a couple of first principles. Number one, AI is being used whether you like it or not, and therefore you ought to seek approvals for the use of the AI that most of the enterprise is using. And so if you talk to your colleagues and see what they're doing with AI to solve the business problems they're trying to overcome, that's where you can get some synergy and you can pick some low hanging fruit.

The typical, um, low hanging fruit in the legal side of things is a non-disclosure agreement. Companies in enter, enter into contracts all the time, and most contracts require an NDA. How can you automate that or set forth protocols on the intranet so that you can give the autonomy to the sales team or the procurement team so they don't have to involve legal.

And so the benefit of AI from an enterprise perspective is automating what used to be multiple cycles of reviews, often involving the legal department. And so one of the things that I recommend enterprises do is look to see what they're using now. And that could be Microsoft 365 that has copilot.

It could be Google, it has Gemini, right? Look at what you have in-house. Now that typically is pretty secure.

It could be tied to your private cloud like Azure. We have Microsoft 365 tied to Azure. And that mitigates a lot of the concern that the legal department will have, which is data security.

Data confidentiality is, is what you're taking onto the machine? Is it looking at data that is very sensitive or personal or customer related? And if so, is it trading on that data?

Better not be, unless it's an internal solution. Um, and how secure is it? What is the confidentiality?

What are the protocols? Um, is it on a, you know, enterprise server, is it in the third party cloud? Those are the sorts of things you have to look at once you pick a solution that solves the problem you're trying to overcome.

And so it's, you, you just sort of answered, uh, my next question, which is, you know, what are surefire ways that, um, a wonderful startup company who wants to do business with your company, um, what are some surefire ways that they could get sideways with legal and have that effort crumble into dust? And it sounds like not paying attention to things like data security and privacy, not being clear about, um, say the difference between retrieval, augmented generation and fine tuning via training. Um, those are sort of, those are things that they're going to have to, uh, I mean this is, this is sort of like, uh, you know, when, uh, when my daughter would, uh, bring home someone who she was going to go on a date with, uh, in high school and I had, and they had to meet my approval, um, there were things that I would immediately leap to as, yeah, this is not gonna happen.

And uh, it sounds like a lot of it is data privacy, security. Um, what, what are some, what are some other things? What's the worst thing that I could say to you if you are the general counsel that's been, that's been brought into the meeting to decide whether or not your company is gonna do business with me, AI startup or ai big company, um, what's a surefire way?

What could I say to absolutely destroy my opportunity for doing business with you, that you would guarantee security If I tell you I can guarantee security? Yeah, I've heard that before, Dave. I love that.

So you, so, so you're saying, wait a minute, you're saying that, you're saying that, uh, that, that attorneys prefer honesty and openness. I'm, yes. Based upon past practice, you know, that's not the case.

And so what a vendor comes in, obviously credibility is key. Yeah. And what, what we try to do is work with vendors who we know.

What does that mean? We know the principles, right? We know people that are working there that we've worked with before, or if I don't necessarily, uh, the CEO does or the board does.

And that raises the accountability both internally and with the vendor, because I will say something somewhat controversial, um, as a lawyer, but contracts are meaningless. What, what does that mean? That means if an issue arises, if you're reaching for the contract, you've already lost, you should be reaching for the phone and contacting the vendor.

And so you wanna have a relationship with whomever you're gonna enter into a mission critical contract with, and that's what I consider ai, obviously. And so, you know, do you have a relationship? Uh, what does your network recommend?

You know, you should do some diligence vis-a-vis your professional network conferences, third parties, Reddit, YouTube. Right? What are other people saying about the solution you're looking at?

Uh, why reinvent the wheel if somebody's already done it? Um, the other thing that would cause alarm bells for a vendor is, you know, if I say, here is the problem that I'm trying to solve. Yeah.

They not only say we can solve that, but we can solve a whole host of other issues. Meaning they're trying to sell a lot of bells and whistles. I'm just trying to solve that one problem.

You can't. Yeah. But you can't, you can't fault them for that if they know that you're, you know, you come in asking for a burger and they're gonna say, well, we have delicious fries also.

Um, no. Agreed. And, and they're in business to obviously make a living.

Right. And I'm happy to continue that relationship, but let's start at the ground floor and Fair enough. Let's get an NDA in place because I'm going to give you not a beta test case you give me that's used with other customers.

I'm gonna give you my use case, right. That will solve the actual problem I have. I'll see how you do on that.

Yeah, it, it's, it's interesting because, um, I think that you highlight the idea that, um, lawyering within a business, within sort of the purview of the general counsel is very much a business focused activity. And, uh, and so with that in mind, I wonder when we start thinking about things like intellectual property law and, and, and things like that, are you from the GC office perspective in a, in a, in an enterprise? Are you leaning on your vendors and suppliers to sort of own that issue?

For example, if, if I'm a, if, if I'm bringing you a solution that you're going to pay for, and that solution includes the generation and retrieval and synthesis of information that creates something that then you use, that someone could look at and accuse you of violating some copyright or trademark or intellectual property, you're, you're really relying on those vendors to take care of that stuff, right? That's not, you don't consider that to be your job. Am I, am I reading that correctly?

Uh, you know, the general counsel's not gonna have people focus specifically on, on, um, on sorting through those tools. Am I, am I reading that correctly? And you may, you are reading that correctly, and it's a great reality check for your podcast name, because even if I have that internal, um, IP engineering legal talent, I may not have the time.

And therefore it gets back to credibility and track record and what you're talking about, the legal term is indemnification and the enterprise better be darn sure it's got indemnification in the agreement. And like a lot of ai, it may not be full stack, it may be a wrapper, and therefore you have to unpeel that onion to get indemnification from the third party that the vendor is relying upon. Well, does that, so does, does that idea of indemnification make it harder for, uh, smaller entrant in this space?

So for ex, so let's, let's say that, let's say that I have a legitimately interesting, valuable, at least piece of the solution that you need. Is it more likely that you're going to be comfortable doing business with me through a larger partner that I partner with, who you feel has, uh, for lack of a better term, deeper pockets, more, more survivability power moving forward? Uh, more of an ability to weather any legal onslaught over things like intellectual property?

Is it, it, does it make it tough for some of the, the smaller startups because you just don't have the time to evaluate their credibility in the same way that you might have a level of comfort with someone like a hyperscale cloud provider, um, or a big model purveyor or a big, a big, um, look, a global systems integrator, a big OEM of of, of, of equipment. Does it make it harder for these smaller folks? That was a very long way.

It does very long way to ask that question. It does, but yeah, it's a cost benefit, right? Yeah.

But it's getting harder for those smaller folks every day as these foundation models get into the application layer. Right? Anthropic just announced the legal plug plugin on cowork.

Yeah. And is it, you know, as perfect as some of these legal AI wrappers as far as functionality? No, but it's getting there and for a lot of enterprises, it's good enough, uh, because you still are required to have a human in the loop at the last mile of any legal review that you're doing, leveraging the technology.

And so that's one of the cost benefit analyses you'll do is a startup got some really great functionality, um, with m and a due diligence where you can't get that from anybody else. You've got a deal coming up, it's gotta close before the quarter end, et cetera, et cetera. Um, you do the best you can contractually, but again, that's just a contract.

You wanna have a relationship. And the most important element, um, is trust. And so do you really trust the vendor to do the right thing when the inev inevitable may happen, right?

There's a data breach. Are they gonna tell you before you discover the issue? And it really depends on what kind of data you're giving that vendor and where it's going to reside.

And so there are ways you can mitigate the risk, but if you have to go with that small vendor, one of the things you'll do, Dave, is you'll negotiate a shorter term contract without, and you'll have payments that are tied to implementation KPIs. Um, okay. You know, clean soc reviews, those sorts of things so that you're structuring the contract to mitigate the risk that you're concerned about.

So are you do, do, do open AI and, uh, and or anthropic, um, by virtue of how well they're known and what they've done so far, do they rise to the level of, of someone that you would see as credible as say, an AWS or a Google Cloud or an Azure or a Dell or an HPE or, or someone like that? Or are you, or, or, or, or do you, and you know, folks like you look for them to be partnering with legacy SaaS folks. We just recently came off of a situation where the stock market reacted to this idea that some latest tool that was released is going to destroy all SaaS because now people can just build things on their own.

I think it was a bit of an overreaction, but, but is it, are, are you, you don't have to tell me specifically whether, but, but I mean, the level of comfort with an open AI or a, an anthropic coming in, um, let's be real. They're very, very new to this whole engaging with Enterprises thing. I see them going the partnership route, which I think is the right route working with SaaS providers.

But, um, are, are open AI and anthropic at a point where they would, they, they could gain a strategic seat at your table? Well, I have my bias, and we don't necessarily have to get into that, but I, what I would look for a vendor that focuses on the enterprise as opposed to the consumer. Okay.

And I know both are, are changing their go-to market models and they're trying to get into both, uh, swim lanes. But, and by the way, I, I didn't mean I didn't, I didn't mean to peg one against the other, I just mean that's okay in general, but No. Gotcha.

Okay. Yeah, yeah. But as an example, you know, a lot of the wrappers are on top of these foundation models, and the wrapper will say, we don't train on your data, or there is zero data retention.

And that may be true vis-a-vis the wrapper, but it's not true based upon the foundation model, especially if that foundation model gets sued and has to provide data and discovery. And that's actually happened. Interesting.

And so is zero data retention or ZDR other acronyms? Is that real? Right, right.

Yeah. I don't know that there's any such thing as zero data retention in 2026. I think we'd be, I think we've, we've uh, we've evolved beyond anything that is created ever being destroyed in terms of data.

Um, you know, I I think that I became educated on this point, ironically, uh, considering what I do with, uh, the AI program at Wharton and other places. Um, I think you were the person who made a point in, in a conversation that we had offline, um, about this idea of, uh, where generative AI fits in the legal profession. And I think it was as simple as a comment of like, oh, no, no, no, generative ai, and we don't use generative AI for, for legal stuff.

I'm like, what do you mean? And it's like, well, of course, no, for research, it was for legal research. No, no.

Generative ai legal research. No, those are incompatible. And, and of course, the, the reason why is obvious when you think about it, which is no, if you're looking for some case or precedent, it has to be letter for letter.

The, it has to be a citation. I don't know all the legal terminology, but you can't just make up an approximation. I can't say, well, you know, you know, the fourth amendment is something about it kind of has something to do with, it's like, no, no, no, no, no.

I need you to tell me what the text is and then tell me all of the precedents. Exactly. Those are things that have to be retrieved, they can't be generated.

Um, how often are you faced with situations where folks in your enterprise don't understand the difference between information that's being generated versus information that's being retrieved? Is that still a problem? It still is.

Uh, we're talking about hallucinations. We're talking about two things, really, if the AI will hallucinate, which it does, and we're also talking about being intellectually lazy. And I'm not aware of too many in-house counsel that are intellectually lazy, definitely not on my teams.

Um, but you have to check the outputs and the models have gotten better, and there's a lot of great, um, legal AI out there that mitigate that risk. But you're still required both ethically and practically to check the output. And there are ways of doing that.

But there are still every week, if not every day horror stories of litigators relying upon legal AI for citations that are just not true. And more and more courts are holding them in contempt for that. And so AI is the latest technology.

It really doesn't change what in-house counsel is required to do, which is not be perfect, but to check the work. And when you're dealing with citations at a legal brief, that standard is pretty high. Um, day-to-day in-house, when you're not dealing with litigation, but you're dealing with trying to get stuff done, um, not perfectly, but well enough to beat a particular, um, milestone, um, you know, you have to have a mitigation plan so that if things, you gotta be able to look around the corner.

Well, okay, the company decided to go a different path than my recommendation, as long as it's not illegal, I have to salute and get it done and put in place a mitigation path. And a lot of that mitigation deals with having a safety net. If something happens, like if there's a data leakage, do I have an incident response plan in place?

Because it's not a matter of if, it's a matter of when. And the biggest risk with something like that happening isn't the breach itself necessarily. It's how you respond.

And so before the, uh, fire happens, you have to prepare for what you will do, what it does. And so you have to work with the technical teams to understand the functionality and what could go wrong and how best to mitigate at risk it's uncharted territory. Um, a lot of the companies, including mine, were regulated either directly or we had customers that faced regulations.

And so one of the best things you could do to prepare for a regulatory issue is getting back to what ai, AI can't do, which is develop relationships with humans, go meet the regulators and develop a relationship with them, one based upon trust, so that if and when something happens, you know who to call. And there's a, a relationship there that you can rely on to help get through the issue. So let me gauge where your head is at with this right now.

Um, bill, have you signed over power of attorney to, uh, an AI agent? I have not. I've used agents.

Do you, do you, do you think you'll, this year? No, but I did sign up for Run a Human and, uh, I'm providing fractional GC service. Okay.

No, I think, um, it, it was Jensen who, you know, was asked as they all are now, you know, is AI gonna replace jobs? And he said, no, for those who know how to use it. And I believe that's the case with And hows council, I think if you can use it, you become more strategic in the enterprise because what the CEO really wants from his or her lawyer, is to be the consigliere, meaning judgment.

Okay. The tool has gave me this great summary that would've taken you price an hour to do it. It spit it out in three minutes, and it's very clear and it gives a recommendation, but I'm still gonna rely upon you and your judgment, and I'm gonna look you in the eye and say, okay, this is what the machine says, what do you think we should do?

Yeah. And that judgment, that business acumen, that strategic value, I think will be enhanced by the use of legal ai. And that's the real benefit to in-house counsel.

But whether it's, whether it's, uh, in-house counsel, uh, anyone on the legal team or anyone else in the organization, ultimately the person wielding the tool is responsible for the damage they do with that tool. And it, it feels like that isn't going to change, uh, at any time in the near future. We're not going to be able to fine nor imprison, uh, AI agents no matter how anthropomorphic they become.

Uh, would you, would you struggling with that? The buck is, the buck is stocky stopping with the biological entity for as long, as far as we can see into the future? Yes.

It's one of the challenges right now with public boards. They need to use ai, their investors want them to become more efficient, and yet who's accountable and what's the governance structure? And a lot of the boards are struggling having a director that has AI experience.

And so a lot of the boards now are looking for people that are AI literate, AI first, that have experience with governance, uh, such as someone like myself that you know, and, and long ago, like six to 12 months ago, um, you know, GCs may not be the first choice for a board, but a GC who's AI literate now is very much in demand, or I think should be because of the governance questions regarding this incredible new technology that is moving faster than the speed of light. I wanna go back to something that you, you mentioned, um, I asked you, you know, what would be, what would be the worst thing that I could say if I was seeking to do business with you and how would be, and one of the things you mentioned, it was, uh, the idea that, you know, if I claim that everything is a hundred percent secure, uh, and the other is that if I'm trying to kind of take up, you know, instead of focusing on solving the problem that we've identified together, uh, that I tried, that I immediately drift off into other things before I've actually demonstrated to you that I can deliver. Um, talk to me a little more about, about that, about that aspect of, um, I mean, are you saying that you would prefer working with someone who was open about the idea that, Hey, we have a lot of experience doing these things, like what we wanna be able to help you do, um, but this will be the first time that we've solved a problem exactly like yours because your problem is unique.

Are you okay with hearing that from someone, or do you need someone to come in and, uh, and uh, and sort of woo you with delusional optimism and the implication that they've done everything before and they've solved all problems? Are you okay with vendors truly being partners in solving problems with you? Or are you like, no, I am not willing to take the risk unless this person's done it 10 times, even though no one has done it 10 times, which is the dilemma in this space.

Uh, I, I, I frankly see a lot of people who, uh, overstate their, uh, level of experience because they're afraid that the prospective client won't do business with them unless they perceive that they have this experience. But the reality is no one does. So are, so were you saying that you're comfortable with people admitting their limitations and exploring the holes that might be in a solution to mitigate against damage that might happen?

Yes. As you said, I don't think you have a choice, but this technology, yeah, because the problems aren't necessarily unique, but the way that AI would solve them could be because like any new technology, there are legal issues that haven't even been considered. And if it gets back to, like, I use outside counsel, I don't hire a law firm.

I hire a lawyer at that firm. I don't hire a vendor. I'm gonna hire the principal, the implementer, the partner that's gonna be with me through thick and thin.

And so that's where the relationship is key. It's kinda like Ghostbusters, right? When there's an issue, who are you gonna call?

Yeah. And so I don't think we have a choice at this current stage of having perfect foresight, um, with how the models are going to work or handle your unique case. Um, that's where I think you can test it out in a secure server, a private cloud, right?

I, I strongly recommend if you're dealing with third party sensitive data, that's a must. And so before you go to production, you have to test and retest and make sure you've got backup, uh, and adequate security. And it's really a cross-functional, um, effort.

You don't wanna become a bureaucracy. It can't, you don't have the time, but you need to have, you know, people in each function that you rely on internally. Um, and if you've chosen the right vendor, then you have the person at that firm that will be there, um, when there is an issue or when there's a question that can't be answered either internally or with the vendor.

You know, you have to be able to work as a partner, like you said, to get through that. And oftentimes the vendors I ultimately choose are the ones that say, we can't do that. Gotcha.

So this, so, so what I'm hearing, what I'm hearing here, uh, as we, as we wrap, uh, is, uh, if you're, if you're a vendor seeking to become a partner, uh, with enterprises, don't, don't try the old, fake it till you make it. Just don't do it. This is all way too important.

Uh, make it till you make it and, uh, and, and partner along the way to create these solutions. Um, I think that goes really far. And I, I, I frankly advise people, um, that, uh, when someone tells you that they have all of the answers, you should run, uh, in this space because no one does and things are changing so quickly.

Bill, thanks so much for this reality check on the world of the General counsel and how General counsels work with their enterprise business partners to make decisions in this space. Thank you, Dave. It's been a pleasure.

Really enjoyed the conversation. Look forward to continuing it. Thank you very much.

I suspect that we'll be having conversations in the future about this as things change for reality check. I'm Dave Nicholson. Thanks for joining us.

Hey everyone, it's Shimmy and welcome to my second Shimmy says this week, if you didn't catch my one yesterday about Apple being the AI of choice for the Edge, check it out. It's a great one. But today I wanted to talk about something else, and unfortunately it's a little more serious.

I think, you know, I wrote over the last couple weeks I've written several articles about csa. It started with when they took their ball and went home because Jen Easterly was made the CEO at RSA conference, RSAC. But beyond that, I mean, that was just toddler like petulance.

But beyond that, the, the problems there are deeper. The first one I wrote was called CCR on Life support. And the basic point behind that article was that the agency that is tasked with helping defend America's digital infrastructure looked like it was slowly being hollowed out from the inside.

Their budgets were cut, their people were cut, they were reassigned, budget was moved over, leadership was let go vengeful kind of political stuff in, in the way of not, not, not keeping the right people, not the kind of sh stuff that maybe shows up in a dramatic headline, but the kind of stuff, it's like a slow institutional death. 'cause it catches up to you, the people, the people are the heart and soul of the organization. You know, a few days after that, I wrote another piece that things were worse than we thought, right?

Because when they let Jen Easterly go, they never had a permanent replacement put in and that they finally nominated someone, but he was being, he wasn't replaced yet. So they had a temporary, uh, direct to put in. Well, this article and the article that I wrote there was, based upon what came out with that, that, you know, the public story we were talking about was just part of the mess.

Internally, things were even a bigger s**t show, and I'm sorry to say it, but a s**t show, you know, turmoil, confusion about direction, morale issues, institutional rot that makes people nervous when they actually realize what the mission is over there. And then this week we got another story. So they fired that temporary director because he really was a clown.

And they, you know, the guy who's been nominated and has not been approved, and then he was renominated, he's still not approved. Well, CBS news reports the other day that, you know, while he was waiting to get approved, he's been, uh, appointed a senior advisor at the Department of Homeland Security working with the Coast Guard. Well, it seems he was escorted out of the building yesterday and dismissed from his position.

No one knows why, but you don't get escorted out of the building with your dismissal unless something, something's going on. And according to the reporting, his access badge was taken away. So not only was he escorted outta the building, he was told, don't come back.

Now, look, they haven't given no one, they've been mom on what this was all about. So we don't know what really happened here. We don't have an explanation.

And I'm eventually, I'm sure we'll find out more of the details. But looking at what we know right now, you gotta wonder, is that the person you want running csa? And how much longer is he still gonna be the designated, uh, supervisor, administrator for csa, the government, you know, the administration still still insists he's their man, but if, if this happened to him, is this really someone you aren't running csar, especially on top of what's gone on here lately since the, the, the, the firings, the layoffs, the budget cut, the mission scope change, you know, 'cause being walked out of a federal facility and having your credentials pulled, it's not usually how a normal job transition works.

I think you have to agree. It's the kind of things that tend, it's the kind of thing that tends to raise questions. And guys, like I said, we've got enough questions, but the biggest one is an obvious one.

What the hell is going on at Cesar? It's like a circus worse than a circus. It's like a clown show.

This agency's already dealing with enough problems. The leader leadership situation has been unstable now for a long time. Budgets have been cut, people have been transferred, other good people have left with the government shutdown, there's like a thousand people there now.

The workforce is still shrinking by all accounts, morale has taken a huge hit. Is this really how you wanna run the central cyber defense agency that's tasked with taking care of critical infrastructure on stuff like what's going on there Now, continuity experience, relationships between government and industry. People who know how systems work and can pick up the phone during a crisis of movings and get things moving.

You don't build that overnight. This takes time. It takes culture, it takes people.

And when you lose it, like they've been losing it, rebuilding, it can take years, I'm afraid. So when the nominee to run the agency suddenly, suddenly shows up in the news because he was escorted out of a building, people should take notice. You know, they say what they mean, they mean what they say.

Even if this story does eventually turn out to be something else, the optics are just not good. There's another piece of the puzzle though, that we gotta think about. And I wrote about it as well this week.

We're at war. We're at war with an adversary who has invested for asymmetrical, uh, cyber warfare. They are desperate.

They're, they're getting pounded every day. If you don't think they're gonna unleash the dogs, and we're gonna see cyber attacks and cyber terrorism, you're crazy at the same time, right? When we need our federal cyber defenses to be working hand in hand with private industry, what did they do?

They boycott RSA 'cause they had a hissy fit that Jen Easterlies over there. An event with tens of thousands, 40, 45,000. Every vendor practitioner you could think of, no interaction with the government there because they took their balls and went home.

It's not the way the cyber cybersecurity community works. It works by collaboration, by working together. Government itself can't defend the internet and industry itself can't defend critical infrastructure.

Both of them have to work together. And I, it's not happening at CSA right now. So all of this has raised a lot of eyebrows.

When it becomes clear that csa, the FBI and the NSA are basically taking their ball and going home, not showing up the way they normally do, not interacting, this is, this is highly unusual. It's highly dysfunctional. Now, as I mentioned before, we're at the middle of a war with Iran that has cyber.

There's so much going on in, you know, the geopolitical moment here. Is this really the time we want CSA to be a circus? It, it, it can't be.

We need to step back here and look at all of this, you know, mil this military kind. They, we sh we sunk a ship in Sri Lanka by submarine. The whole Middle East is being attacked.

The that region is volatile. Iran's not the only cyber terrorist capable organization. They have all the proxies in Iraq and Lebanon and the Houthis.

And, and make no mistake, Iran has spent years developing these cyber capabilities in the past. They've already targeted banks, they've targeted infrastructure. If you don't think they're gonna do it this time, I don't know where you know, where you're getting your information.

'cause at the, the, at the end of the day, it's probably one of the only avenues they have left for an asym asymmetric, uh, asymmetrical attack. They don't have much left. So as things become more desperate, desperate people do desperate things, and we're gonna see desperation in cyber attacks, it's exactly the time that we need cisa, right?

It's, so this is, this moment's a particularly bad time for leadership and chaos. Bad morale at this, at America's cyber defense agency, which is what Cesar is. This is exactly the time we need stability.

We need people who've been at the helm to be back at the helm. It's when you want the private sector to be confident that they have a stable partner in the government side of it that's working smoothly. That together we can weather this potential attacks from this enemy.

But instead we have confusion. We have a circus, we have budget fights, we have leadership drama, we have innuendo, we have morale, we have just craziness. We have a nominee who just was in the news for being warped out.

The three major cyber organizations stepping back from RSA, this is like a perfect storm of dead crap going on in normal times. I'd say it's, it's worrying. But in today's times, I'd say this could be a catastrophe in the making.

Let me say something though. The people inside Seesa, the ones who actually go in and go to work every day, even with this government shutdown, you know, they're good people. Some of the best people, they left their private sector jobs to serve the country.

They're talented people who believe in the mission at csa. And in spite of all this, they show up every day trying to keep things running. But you can't run these institutions on fumes forever.

And on the dedication of the staff alone, they need leadership. They need resources, they need stability, they need, they need supporters. And they're not getting it right now.

It looks more and more shaky. And, and, and here's another thing to remember. Do you think the criminals, do you think the Iranian cyber attackers are saying, ah, let's wait for Washington to sort itself out.

Let's wait till they appoint a new director. No, they don't pause or wait. They see a weakness, they see an opening.

And this is, and we're vulnerable. Critical infrastructure operators don't get to delay attacks until the organizational charts finalize guys, these are real constant threats that we gotta deal with now. And that's why this current situation, you can laugh and say send in the clowns, but it's frustrating and it's a potential catastrophe waiting to happen.

The pol politic politicalization over csa, the budget fights the chaos. Well, it's all part of what Washington is today. This administration thrives on just creating chaos.

But chaos doesn't work when the world comes crashing in. We hope. I hope you should hope that the professionals who are still inside there can keep this system running.

That they can continue to coordinate with the private sector. They can continue to defend the infrastructure that our country depends on. Because if the instability continues much longer, there's not much hope there.

We may eventually learn something or lesson here the hard way. Now when the circus comes to town and they set up the big top at the top of the nation cyber defense engine agency, it's really isn't time to send in the clowns guys, not a clown show. Anyway, that's it for this episode of Shimmy Says, I hope you en enjoyed the whole week worth of shimmy text on gang.

Hey, if you want the full breakdown on this, I do have an article up on Security Boulevard called The Circus at CSA Continues. I'll see you next week on Shimmy says on the gang, enjoy your weekend. We're out.

Let's stay safe. My name is Megan Kuchi. I'm part of the AI infrastructure networking group at Cisco as a technical marketing engineer.

And today virtually joining with me is Richard Leko who leads this entire team. And let's face it, as much as we are obsessed with different models as engineers, we know that the real battle is at the infrastructure layer. And that is what exactly we are going to talk about in terms of how Cisco is trying to simplify from an infrastructure perspective.

The cluster designs, the automation, the end-to-end visibility so that you're not finding needle in a haystack sort of a problem when your G GPU take say months to design weeks to troubleshoot, we are not losing that time only. We are losing the competitive edge. So that's what we are trying to emphasize on for this session.

And before I get started, I just want to make a quick call out to previous session on networking field day 39 where we went in detail and Paresh did a fantastic job of covering all the AI cluster design components, the challenges from bits and bytes level, all the way to congestion control mechanism with dynamic load balancing and cabling plants and everything. You can refer to this for more of a deep dial, but for today what we are trying to go ahead and cover is more of the reference architectures. How do we build this AI cluster and what are the key components tied with this AI cluster?

And at the same time we'll try to look at a demo of what new updates we have made with our Nexus dashboard platforms and how we can design automate right at scale and provide that end-to-end visibility from this dashboard. After that, I will try to hand it over to the hyper fabric team. We'll try to focus on more of the cloud managed piece for it.

So with that open to any of the questions because we want to keep it as interactive as possible. And let's get started with Cisco AI networking. Now being part of this technical marketing engineering team, we get to interact with lots and lots of customers across different segments.

And after all these AI infrastructure boom, we help key customers who are hyperscalers, who are scaling from tens to thousands to millions of GPUs which are out there. They have separate needs, separate, uh, you can say expertise and everything which is out there. When we talk with new cloud vendors, what the GPU as a service vendors, they are building thousands and thousands of scale in typical multi-vendor environment fashion.

And then we are seeing this uptick in terms of enterprises that we are targeting and talking more and more in terms of universities, automobile manufacturing plants. And I will give you the ex actual examples of the customer deployments where we are designing. But those designs might be just tens or thousands of GPU U that they are looking at.

And how does Cisco AI networking fit into it? Number one thing that we have as part of our three unique pillars is the systems. We build our own custom silicon.

Silicon one platforms if you're familiar with it. And sometimes we get asked the question why not rely on merchant silicon? But the key differentiation that we have with this is that with ai, the infrastructure requirements are all the time changing rapidly.

And we are working in lockstep for, with this large scale customers, to make sure that we go ahead, leverage our programmable pipelines and make sure the features that they're looking for are available without re-spin of asics and then go ahead and support, start supporting it on our platforms. At the same time, we have this partnership with Nvidia where if customers are looking for that full stack reference architecture compliance, you can have the same Cisco switching platforms running the Cisco NX OS as the customers are already having NX OS deployed in their uh, data center fabrics, but have the Nvidia spectrum X silicon running on top of it. And we will talk about that as well in detail.

We have our own transceivers, which are rigorously tested. The key benefit that you get is if you're connecting with NVIDIA Connect X seven eight N or a MD Polar, we have these transceivers validated as well as checked at the unit level testing, uh, stress testing, entire performance testing, everything is done for these trans Cs. And we have tried our own NXO software, which is across the data center mature platform now being efficient for AI workloads.

The second important pillar that we have is the operating model. We have Nexus dashboard, which is for on premises management and for full stack cloud managed solution that you want, you would have Nexus hyper fabric. At the same time, when we talk to new cloud customers hyperscalers, they have that expertise where they do not want to rely on these management vendor specific management dashboards.

They want to integrate with their own APIs. And that's why even we take this API first approach to have those APIs included as part of their automation framework. And third important thing is that you would see lots and lots of AI reference architectures, which are, which are validated design architectures with performance benchmarking teams along with all the different vendors that we try to support, which is Nvidia, A MD, Intel, all the storage vendors like TD and CCA and so on.

So the key thing is that we want to make sure that we have consistent networking experience irrespective of what vendor you are interacting with. So this is exactly where we are with our AI networking story and we are constantly expanding and enhancing on top of it. So for the first piece, what I will try to do is focus on the systems and reference architecture to understand that what we are trying to solve over here.

Now, first of all, we need to understand the definition of AI cluster because sometimes whenever we talk with customers as well, AI cluster is nothing but GPU to GPU backend communication network, which is east west traffic. Mm-hmm. Well that's not true when you are trying to design the network.

Actually it's more than that. So say for example, if you have your GPU nodes which are out there, the first and foremost thing that you would require is to make your GPUs accessible to users, applications, services, schedulers. And that's where front end network would come into picture.

The traffic pattern over here would not be that much bus. It may be moderate in comparison to East, west GPU communications, but at the same time your inference traffic runs on this front end. So it'll be bursty in that capacity, but at the same time, latency sensitive on top of it, you would have storage network, which can be high speed storage leveraging say vast cca, DDN and so on.

And historically these two have been separate networks where due to different QS profiles, latency profiles, this have been running separately. But more and more we are deploying an actual customer, uh, scenarios. We are seeing that they're converging both frontend and storage because of more and more 400 gig, 800 gig proliferation that we are seeing.

And we are seeing that front end traffic is anyways limited or less compared to the storage traffic that you would have from a high bandwidth perspective. Megan? Mm-hmm.

Uh, Jack Poller with Paradigm Technica. So you earlier mentioned you're talking about the, the type of traffic for inferencing. Do you make a distinction in the architecture between a cluster targeted at inferencing and a cluster targeted at training?

Uh, yes. We do have different, uh, load balancing mechanisms and so on. Mm-hmm.

So again, that was covered during the networking field day 39 session where we have different modes in terms of load balancing the traffic, right? And based on the policies and so on. And I'll try to cover during these slides also that how we are trying to have that, well My, my question is more directed at in this discussion, are we discussing something targeted specifically at inferencing or is this more generic?

Well, This will be more generic based on the customer deployments. It can be either training or inferencing and so on. So it's not going to be just inferencing focused.

Thank you. Yep. Alright.

So mm-hmm. Does, does uh, Cisco get involved in the GPU to GPU east west traffic kind of thing? Absolutely.

That's the next thing. So that's where, uh, based on your designs, say if you're scaling out from one node to two nodes to all the way up to say whatever end number of nodes that you have, that's where we have this dedicated backend network. And this is where all the key buzzwords that you might hear about, like lossless, bursty, traffic, long lived flows, elephant flows, lowest latency, lowest jitter, highest job completion time, all those things apply over here because you are doing those collective communication jobs like all to all, all gather, all reduce, which is happening in the network.

And that's where also Cisco has all the solutions for different challenges that you might have in terms of congestion and everything. So this is very sensitive in terms of even minor congestion network delays or anything that might happen in the network because you want to make sure that your GPOs are being utilized at a hundred percent or highest capacity so that you're not wasting your infra investments that you have in your network. So I'm not exactly familiar with the technology, but uh, Nvidia has their own NV link and things of that nature are, are you playing in that space?

Yeah, So right now NVLink and all those technologies are specific within the servers where you can go ahead and scale up your designs, add more GPU nodes and so on for the internal communication that you have. So that is specific to NVIDIA and compute servers. We are not playing in that scenario.

What we are trying to right now go ahead and how as of today is playing the scale out when you want to distribute the nodes across uh, different uh, say GPUs and so on. So like rack to rack versus server To server Server, yeah. Within rack.

Yeah, that's called like a scale up design. And then we have this scale out and I we are also playing now and scale across meaning that YouTube power constraints, cooling constraints and so on data centers like new cloud vendors have data center facility in one location and the other location is also there. We are connecting those via our new asics, which are called P 200 A six and then having it across like hundreds of kilometer distance a cluster which can be a megawatt megawatt cluster together, it can form like a gigawatt plus.

Yeah. So Frederick Van Herrin, uh, ENS consulting. So when I think training I think in finna band.

So are you, is Cisco then Yes. Injecting ethernet to replace in finna band? Correct.

So this is mainly ethernet based solution that we are trying to showcase. And the key benefit that you might get out of this is that if you have separate networks, you are breaking up the operating model, right? And I will try to show you in the demo as well that how when you go with an ethernet route, you can manage the entire cluster as one, add configurations, profile visibility as one as well.

So that's coming up in the demo as well. The benefit you get and then you have so many different components you need to manage your switches, you need to manage your compute nodes, you need to do firmware upgrades, you need to go ahead and have the schedulers and everything pushing down the configurations. This is where out of band management comes into picture.

But that's what the entire AI cluster constitutes of. And this is one of the misconception people just focus on the east west backend network and think that this is what is your AI cluster. That's not true.

Backend network East west can be for your distributed training and inference perspective, but you need to make these GPUs accessible to your users application scheduling logic as well as the storage network which is out there. And the key reason I discussed this AI cluster is because we have dedicated reference architectures which so as a validated blueprint so that you can eliminate all the, so you, you can think about guesswork and get a starting point in how your AI cluster would look like and following the NVIDIA's enterprise reference architecture principles, we have our own Cisco enterprise reference architecture for less than 1,024 GPUs. Where based on your operating model, if you ma want to manage it via on-prem Nexus dashboard or via hyper fabric AI in a full stack form, you can get examples of like 96 GPU cluster, 1 28 GPU cluster, 1,024 GPU cluster along with each and every components.

How many optics do I need? Uh, what switches would be, right? Not just for my backend but even for the front end storage, all those infrastructure which is out there.

So that is the power of this enterprise reference architecture where it sort of gives not just a network diagram but validated certified components which are surely going to interrupt it with each other. So question about that is your reference architecture meant Ken, now I'm sorry then introduce myself. Is your reference architecture meant to be implemented as part of a broader reference architecture from Nvidia say, or compete with it as customers are looking how to, you know, deploy in their environment.

So I I, it'll come up to the NVIDIA versus Cisco reference architecture in the next slide. Okay. So just wait for that.

And this is where again, this is for enterprise reference architecture, at the same time we talk with large scale cloud service providers, the new cloud vendors who are building in capacities of one k, 2K, 4K, eight K, 16 KGPU clusters and that's where we help cloud reference architecture. But this is where we are not only showcasing the backend network, which might be easier to understand, but even the frontend and storage converged network which would come into picture with the Nvidia uh, say HGXH 200 platforms, specx and so on. The examples are there where say there might be thousands of optics, but what optics to use, do I need to have O-S-F-P-D-R eight?

Do I need to have cable lens of five meter a hundred meter? It gives you entire description. So it's not just a network diagram, it gives you different example.

If I want to build one KGPU cluster, how do I do that? If I want to build 4K GPU cluster, how do I do that? What would be the components and everything required?

So coming to your question, right, uh, are we competing with them? No, we are working together and making sure that for the right set of customers we have the right set of platforms. So see if you want to add to the Cisco reference architecture, have that unified operating model with silicon one based asics, which are our custom silicon, asics have your front end backend storage network, everything connected via these switches.

We have the N 9,364 cross 800 gigs switches as of today. And there can be requirements from customers where they want to make sure that they're compliant with NVIDIA's reference architecture in a full stack manner where you might have Nvidia, Nicks, NVIDIA storage, everything which would be compliant with that. And this is where we have, because of this partnership apart from Nvidia Spectrum X switches, we are the only vendor to have N 9,100 series switches with Nvidia Spectrum X with silicon so that you can go ahead and have N-C-P-R-A compliance but at the same time NXOS becomes your common operating point because your expertise, like we talk with enterprises, we talk with new cloud customers, they are already having Cisco operating systems as part of their data center fabric.

They can leverage NVIDIA's silicon, have the Cisco designed with NXOS running on top of it and at the same time manage it via the same dashboard, Cisco dashboard which they're using to manage their current data center fabrics or even their uh, you can say AI clusters that they're planning with. So we are not competing, we are partnering more of more together and making sure that we have right fit of platforms based on customers requirement. If they want to go full n media route, we have N 9,100.

If they want to go with Cisco route, have Nvidia like GPUs, Nick and so on with Spectrum X capability, you have our N 9,300 series. So is that the distinct spray ese from Silver Tank Consulting? Is that the distinction between the hyper fabric versus the non-hyper fabric?

Is that because you're using the Cisco N three and and Hyper fabric, uh, is applicable for both reference architecture? So even uh, the CERA that we showed the uh, I understand that but I mean is it is a is a switches themselves the Cisco N 9,100 versus N 9,300 different between the two fabrics? So right as of today, yes you would have a hyper fabric 6,000 series, but there are plans moving forward where hyper fabric will be supporting the nine Nexus 9,000 series switches as Okay back from hyper fabric.

Okay, so the camera gets uh, with them. So uh, we'll talk a little bit more about this in the second half, half of the session. Um, the operating model is really what's different between hyper fabric but assume that the hardware will be able to uh, use, we use any of those sets of hardware as we build out for both, but we'll dive more into the operating model and sort of why we built it later in the session.

So I guess, I guess the other question is are you limited to using the the 9,100 series or 9,300 series? I mean can you intermix the two? Absolutely you're gonna have your front end network within ninety three hundred and ninety one hundred but at the same time, if you want like a full stack compliance in terms of architecture, you will go with N 9,100 because that's why the customers want that compliance and that's why they have.

But it's running the NXOS operating system so you will be able to have this common experience irrespective of whats which you're going through. I understand the operational experience would be the same. I guess the question I have is like for the backend network yeah you would be required to use the N 9,100 series in order to support that.

Yeah, N 9,100 or N 93. So there is either one. Yeah, either, either one you can, you can support that.

So thank You. Yep. Alright, so I will go and give you an example of how do we design a backend network in an AI cluster and once we get these principles right, it's pretty much straightforward where say example number one thing you want to make sure in an AI backend network is that you have a non locking architecture.

What does that mean? Say for example we are taking 64 800 gig port switch that we have, we cut it exactly in half, not literally, but just in terms of port connections, uh, we make sure that based on the GPU Ns that you're connecting to say if you are having connect X eight NS with support one 800 gig OSFP port, you can connect up to 32 of those connect X eight nicks. And then because you have 32 going to your GPU Nicks, you make sure in terms of uplink you have 32 ports going towards your spine.

This is what a non-blocking architecture means. Or at the same time this switches, if you're connecting to connect X seven nick with 400 gig ports, you can make sure you can use the optics, right optics and the right cables to break it down into 1 28 6, uh, 400 gig ports, which means 64 go down to your GPU mix, 64 go up and we have the highest redx redx meaning that we can even break it down further to a hundred gig ports and have 512 ports that can be powered up from this single switch that we have. So that's how typical designs are based on the customer requirements.

But when we extend it out to say scale out, say if you want to go from one node to other node and so on to have your distributed training or inferencing, one of the critical or foundational design choices that we see is how you connect your GPU nick to the LEAF switches which are there. Mm-hmm And how many of you're familiar with rails optimized design? Alright, I see two of them.

So what does that mean is that when you connect your GPU NIC one you need to make sure that it connects to LEAF one, your GPU NIC two connects to LEAF two and so and so forth. If you are eight GPU NICs in this particular server, so I have eight leafs which are out there, the key benefit that I get is when I try to go ahead and extend this, I make sure each and every GPU is just one hop away from each other. So that say if GPU one of node one wants to talk with GPU one of node 64, they are just one hop away from each other.

And this is how we can power up to say with 400 gig CX seven nicks that we talk about 64, 400 gig ports down link. This is called rails only design. You can get away with it, but what happens when, say Nick one to Leaf one connection fields, what happens when Nick one fails altogether, that's when the traffic has to travel through an alternate path and even in terms of scalability, you require that spine layer for that non-blocking architecture and how that one is to one subscription ratio.

So the animations are a little bit slower. So hopefully I'm not talking too fast, but this is where you would go ahead and have your spine layer coming up. I uh, spine layer coming up in one to one or subscription fashion.

Yeah, finally it's there. So it takes 30 seconds to go ahead and plug cable them and now in 30 seconds we have the ports also and right, so, so that's how we have this non-blocking architecture tied together and you make sure that you have 64 ports from Leaf one going down, you have four spines, so 16 400 gig ports going towards your spine as well. So F Fred and her from ENS Consulting.

So, so I understand the design, so if I compare this with Infinity bands, there's not just the hardware component, it's also the logical component like you UFM has, you know, for Infiniti band. Do you have similar? No, uh, no we do not require that.

We have like right now this particular like fabric which can be configured with ethernet and we can, which we can scale out accordingly leveraging all the ports. So is it fair to say then that your existing ethernet tools are still working? Yeah, Yeah.

Those are working the technologies that we use, configurations and everything, it's beyond the scope of presentation. We covered it in networking field day 39, right? But we have gone through it and you can use exact same configurations and everything that you have in terms of networking com component, uh, concepts and this becomes your sort of a scalable unit.

That's the term that we use, which means that this is a repeatable block and you can just go ahead and keep on adding more and more components or like, uh, like repeating this modular block and you will be able to scale out further. So for example, we are taking this, uh, 64, 400 gig port scaleable unit. The size can vary based on your designs and everything.

This is not a hard stop or fixed solution. So that if I want to go from this 512 GPU 2024 GPU cluster, all I need to do is repeat that exact same block and incrementally upgrade it into scalable unit two. But just to maintain that one to one or subscription ratio, I need to make sure that the links earlier, I had 16 links of 400 gig going through four spines.

Now I have eight spines, so I have eight, 400 gig links. But you can look at how many cables, optics, and everything is going through this, right? Because of this, the reference architecture can give you that cluster bomb analysis where the number of optics, cables, everything can be under By, by one-to-one oversubscription.

You mean it's not oversubscribed? No, it's not. It's just the term that we, I also don't like it.

And whenever I use it, I use one-to-one subscription ratio. But that's what the industry term is like. One is to one over subscription.

Because from data centers perspective, we are always thinking about five to one, seven is to one or subscription ratio. Okay. Well, or or it's one to one under subscription.

Yeah, Exactly. Exactly. And that's why when I try to go ahead and build like a smaller clusters in front of customer, and I'll show you the example that what we do is, I always call it one, is to one subscription ratio that we need to maintain.

So you are absolutely right, but just using the industry from with that, we are, This is, uh, Arian Newsom, uh, I'm from Ethical Tech Matters. I actually have a question about, uh, the GP node failures. So does that hyperscale fabric design prevent, like casca and failures we used to see in the old traditional three tier architectures, Uh, sorry, the question was, Uh, so does hyperscale fabric, does that help prevent some of the cascading failures?

So, so we can identify those failures and we will go into demo with Nexus dashboard and hyper fabric as well, how we can go ahead and look at those failures and try to remediate that which are out there. So for sure, we are coming to that in a couple of minutes. Okay.

Thank you. But we get all the time when we talk with say, not neo cloud vendors or hyperscalers, that, hey, we do not have this much of power. Uh, we do not have these many GPUs, we don't want to build at this scale.

And I just want to share one experience that I had with one of the university customers in East Coast who got National Science Foundation funding for a couple of million dollars, and they wanted to start small. All they had is the budget to have four nodes with eight GPUs each. They want wanted to start small with 32 GPUs and 32 nicks.

And based on all the research and everything for their PhD and every, uh, what, what they had, they wanted to make sure once they get funding for all the universities in that area, they wanted to share this infrastructure. So some of you might be familiar with that, but we were working closely with that. And for them, they were saying that, do I need a separate backend?

Do I need a separate front end? Do I need to build at like eight, leave four spines sort of a scale? The answer is no, because you can just go ahead in this particular design and have two switches just to limit the failure domain to one switch.

You would have two switches. Take those principles for a high end, like high AI cluster scale use, say four ports going from each of the servers to switch one and four ports from G Pix going to switch to, and by the way, the over subscriptions piece I always use when I try to design maintaining one is to one subscription ratio is what we try to go ahead and do over here so that if we have four, four ports going, each means that inter switch connections need to have four ports between them. So similarly for this, like say four servers that we have, we will have 16, 16 ports for the entire block going to the switch and so on.

And then I can just merge my front end storage management network because anyways, these switches are 1 28, 400 gig ports, right? We have plenty of capacity in terms of port density that we have, and we can just converge your front end backend management and even your, uh, the back, uh, storage network altogether. So I'm gonna ask kind of a maybe obvious question.

Mm-hmm. So we're talking about cluster technology, and you just mentioned an example of somebody that got funding and blah blah. Yeah.

So what, what network are we looking at honestly, from, from what stage of the AI lifecycle? Is this helping them clean their data? Is this doing the inferencing?

Is this doing the training or like what does, what traffic flows in this super? So this is your converged network. So for this in university customer, they are mainly doing research work, more of computational analysis and all those things which can be tied to an AI job, which is distributed training or a distributed training for HPC, like high performance computing that they're using.

So this can be mainly from a research perspective that they're using. It's not necessarily inferencing that they're trying to do in most of the cases over here, but for sure they can be looking at once they train the data, how quickly they're able to access the train data with the storage network and so on. So this is more of a converged environment that we are talking about.

So it's, it's, when you say converge, you mean the environment from a technical standpoint, or do you mean it from some usage standpoint? From a technical as well as usage standpoint? So that's what we are trying to say.

So that next time when they get like more funding for adding more GPU servers based on the same principles, say if we have term this as scalable unit one, they can keep on adding based on we can ports more and more servers which are out there and then scale out to more GPUs while Rev leveraging the same networking infrastructure. So, and they can, can they leverage the same networking infrastructure? So as this one set is trained and you have whoever accessing the results to train data, then you can also have, use the same network to do those other two correct funding rounds, but also consistently being able to present that data for people to access along the same network, Along the same network.

That's how we are trying to do that convergence. Mm. Question on that, Brian Martin, uh, signal 65.

So I'm managing a handful of 32 GPU and 64 GPU clusters, just like you're describing here. Uh, when you converge front end traffic storage, traffic backend, GPU traffic, are there any special considerations you have? Absolutely.

Okay. And that was the point I was going to make. I, I don't have the detailed slide because I didn't want to, uh, go into detail conversation and spend the entire right, but that's why we have a Cisco innovation over here leveraging our Silicon one asic, which is our differentiator based on the exact requirements with customer mentioned.

Mm-hmm. And even large scale customers are looking for this when they converge their north, north, uh, north-south networks. So this is what is called policy-based load balancing in mixed mode environment.

So what we are trying to do is, because you would have your storage traffic, your backend GP to GPU traffic, you want to give that preferential treatment, right? So that your GPUs are not sitting idle. If there are, say for example, an AI job is running, there will be, it would be distributed into a hundred tasks.

Those a hundred tasks would be distributed to a hundred threads within the GPU, right? And if you complete 99 threads and one thread is just waiting for the entire process to happen, congested in the path, they're all waiting. They're all waiting.

So you need to make sure that the GPUs get the highest preference rate. Treatment then can be your storage because you're feeding all the data for training traffic to your storage network, and the least can be for your front end network. So that's what we are able to do based on the policies.

Say, say that for my training traffic, GP two GPU traffic, I want to make sure that it gets the best path possible with the highest load balancing scheme that I have. Then for storage can be the second priority, and the third priority can be my front end traffic. Perfect.

So DLB and GLB can run on GPUs and other Products? Absolutely. Yeah.

Excellent, excellent. Yeah, so that's what we are trying to do. So just to be clear, so that sort of supports a, a non-blocking network for the GP to GPU traffic, whereas blocking elsewhere.

Yeah, blocking in the sense that, say your storage can have eight, 400 gig ports based on the storage nodes that you have, and then front end can be anyways, the traffic would be like very lower. So it would be just one 400 gig port and so on that connects to all. Hey everyone, it's Alan Hummel from Techstrong.

Welcome to our next session in our dynamic series of conversations between, uh, select thought leaders at Microsoft as well as the some of the analysts from the Futurum group. In this session, we have Tiffany Tracy, VP of product management for the power platform at Microsoft, and as well as analyst from Futurum Group, Keith Kirkpatrick, the session. This ti this session is titled Agentic Automation.

In this session, Tiffany is gonna lead us on a deep dive into the operational realities of agentic automation. It's a world where apps, agents and chat are converging to reshape enterprise execution. We hope you'll discover how AI empowers everyone with a special focus on those who need accessibility and disability support.

You're gonna learn how business users supervise autonomous agents that execute, escalate, assist driving inclusive productivity, expect insights into multi-agent orchestration, human in the loop governments and chat led transformation across support and product activation. So another great session. Here's Tiffany and Keith.

Thanks, Alan. I'm Keith Kirkpatrick, research director with the feature room group covering enterprise software and digital workflows. Today we're gonna be talking about agentic automation and how it is reshaping enterprise execution where apps, agents and chat functionalities of conversion to assist across workflows driving the external engagement through the delivery of personalized intelligent experiences and streamlining interactions.

And hello, my name is Tiffany Tracy and I'm the VP of product management for the power platform core, which covers our power apps, power automate power pages, RPA and process mining. I've been with Microsoft for 25 years in a variety of product roles. And looking forward to the conversation today, As we're both aware, we really can't get away from a discussion about today's technology without talking about ag agentic ai.

And I wanted to first start off by asking you about some of the ways in which AG agentic AI is changing the way customers are engaging with businesses on a day-to-day basis. Yeah, so I think it's great if we first start with the fact that agentic AI is going to change the way we work, right? We're moving much more into these human led agent operated environments.

And some of the big changes that come with that are, we're gonna move much more from this very task-based focus to a more intent and goal-driven focus. And we're gonna move from working like in a particular app to really working across apps with that we'll see this synergy of humans that are, uh, you know, driving what we're gonna do. They're adding business intelligence, they're guiding, we're gonna have agents that really do a lot of the, the execution work.

We're gonna have intelligent apps where these agents and humans can dock into manage everything. And we're still gonna have automations like we have today for very deterministic workflows. When we put all of that together, what we get from a customer experiences, they're going to get much more personalized and contextually relevant experiences, uh, much faster and with a lot less effort on their part.

And in fact, in many cases, we see that customers, uh, organizations will able to expand the audiences that they can actually serve with this technology. So like a simple example that, that might be, I'm on a flight, turns out I'm gonna miss my connecting flight. You know, today when I land I might get a, a text message that I've missed my connecting flight.

But you see, very quickly I'll land, the airlines has already rebooked me with an agent. They're gonna let me know what my new flight is, and then if that doesn't work for me, they're gonna gimme a human to escalate. That's going to change in these kind of customer experiences.

Can you talk to me a little bit about how we're going to see all of this automation, uh, intelligent automation be managed? So one of the powers of this agentic transformation is you begin to get intelligence on tap. So you have these different agents that you can leverage for different business functions.

A level one agent, I think most of us have probably experienced in this point, and that is AI, is maybe we're asking it questions or it's giving us a set of information. And then you have level two where the human is actually directing the agent to conduct some sort of task. And then the business rules, um, dictate when the, the human will get involved.

And it may be just giving the human information so they can make a better decision. And then level three, three is where you see these agents actually taking action aligned to the business rules and the human being in the loop aligned to whatever business rules you set. So what you'll find is that the goal of how we're thinking about agent AI is we want humans to continue to work in the way they do today.

We want them to have a personal assistant that transcends with them throughout their day, whether in their business data, their productivity data, whatever tasks they're doing, and then they will have intelligent apps, they'll let them manage some of these autonomous agents, but those agents can dock into their personal assistant, they can dock into their agents. So we really want that humans continue to work the way they do today, that this AI will sort of collaborate seamlessly with them. And that's why you see that using both intelligent apps and kind of copilot in this chat interface have their place depending on what the human's trying to accomplish.

And so we want this all to kind of slot in more seamlessly versus thinking about it as like they, they have to change as much the way they work. Right, that makes sense. But I guess one thing that I'm, I'm particularly curious about is as we move into this world where we have agents that work alongside of humans, and there are obviously gonna be agents that work sort of autonomously, obviously still with in a human in the loop to make sure that, that they, that they don't go off the rails.

How do you actually coordinate multiple AI agents across a platform to make sure that, you know, the agents do what they're supposed to do when they're supposed to do it? Yeah, it's an excellent question. It, it's very inherent in, in the platform we're building, uh, uh, across both copilot studio and power platform, and of course some of the pieces in Azure.

But it is very straightforward to design for a particular agent, what its rules are, what it's allowed to do, what knowledge it has, what memory it has, what kind of guardrails it needs to follow. Mm-hmm. And what we see as customers are moving to these level three agents is they're really thinking through their business processes and chunking those up into reusable components.

So maybe for instance, you, you interact to gather information from an external company, and you do that for several business processes. You might build a dedicated agent that does that and and gathers that information that will have a set of business rules that you set for that agent. It will have a set of points where you escalate to a human or where the agent can actually take action, and then that agent may talk to another agent.

Again, you define what that communication is and the business rules. So it's very configurable to what your business policies are, what your risk tolerance is, depending on the, on the impact. The other piece is it's quite straightforward to evolve those business rules.

So maybe for instance, you start with an agent that makes recommendations on approving insurance claims or approving purchase orders. You might say that when you start, every single one of those has to be validated by a human. Then maybe you say, wow, that's going really well.

If it's, you know, under such amount a thousand dollars, the agent can auto approve if it's over that the human still has to make that decision. And then you keep ratcheting that up as you build confidence in the, the agentic system you've created. And those things are very straightforward to configure and continuing to evolve Actually.

How does power platform help to sort of manage that, that, as you're talking about multi-agent orchestration across different modalities, whether we're talking about chats, uh, applications and backend systems, because that seems like that's gonna be a core sort of, uh, requirement as organizations, whether they're dealing with regulated industries or not. Absolutely. So when you think about the power platform one, we, we have a tremendous amount of line of business, large scale apps running on the platform today.

And I think it's really important to note for those customers, we are going to bring AI to where they're working today and let them use AI to add even more value to the, the applications they have today. Then we're introducing new tools, uh, for building agents and some of these intelligent apps that will, will dock the agents in. All of that will still run on the power platform managed environments.

So all of the governance that you're used to in the power platform will extend to this agentic transformation so that customers have confidence that they are running in a managed environment, that they have the ability to set the policies to manage it, to audit it, to understand RAI, all of the different components they need. But that will be within the, the core platform that they have come to, to trust in in managed environments. Yeah.

Tiffany, you just mentioned something that's really interesting and, and you've been talking about it throughout our conversation about the idea of human in the loop governance. I'm curious, how do you actually embed that into agentic workflows without sort of slowing down automations or creating unnecessary bottlenecks? So human in the loop can be orchestrated at any milestone in the process that makes sense for that process or that business.

This is one of the places that we think intelligent power apps is going to play a large role. So you can imagine that I might have, you know, a thousand automations or a thousand agents that are running, and I have this intelligent app that lets me go through and quickly approve, guide, change, whatever needs to happen to ensure that the human is guiding but not slowing down the process. And I think this is one of the roles we see for intelligent apps as we go forward.

What about, you know, the other thing I've heard about is the use of adaptive risk models and how that might help ensure that agents just remain compliant with any kind of regulatory or even indu or even, uh, business guidelines. Can you talk to me a little bit about that? So for every agentic solution, the organization really needs to think through a concept we call evals.

And those evals are what are letting you know that the quality, the functionality, the reliability is all within your guidelines. And so it depends on the agent solution, but you are going to have metrics that tell you the functionality and the reliability. It's gonna let you know the quality of the response.

If it's a agent that's creating some sort of UX or interface, you're gonna have metrics that let you test if that is, is high quality and functional. Um, and then of course you're going to have, have evals around responsible ai. And so depending on the solution, one of the first things you want to do as you get started is define for the type of solution you have, what are the areas that will be key and what are the metrics and tests you want to use?

And then there'll be multiple ways to ensure that those metrics are on track. So we've heard a lot about hi agentic ai, but one of the things that I hear from talking with companies is that there's still a little bit of fuzziness or confusion around what sets, uh, agentic AI apart from some of the chatbots or assistance that we become, become accustomed to dealing with in our everyday lives. There's a number of things.

One is that an agent, if you give it to them, has memory so they can remember previous conversations with you. They can remember previous context. The second is that the agent can learn, you can continue to train it on knowledge, and it can continue to learn and help be more and more helpful as it goes along.

It also has not just the initial, uh, knowledge that you trained it on, but it has generative ai, which helps it to fill in the knowledge that you've given it. So you can think of it has all the power of the, the orchestration and the LLM or the larger language model with your specific information on top to personalize it. All of those are things that chatbots could not do.

Chatbots also cannot take action. So chatbot was really, it was a great at the time, but it's really more of like a q and a with very curated answers. When we get to LLM, it has all of these richer capabilities.

And so it's not only quicker to get the information back to the human, but it also can do more of that on its own because of the context, the shared memory, the knowledge, and the fact it can take actions. Well, one of the things I think that age agent AI is really sort of building on is that chat modality where you're able to use natural language to interact with it. Uh, do you see that as being sort of, you know, another sort of real selling point for using age agent ai?

Because you are able to, you know, anyone can interact with it. You don't need to have, you don't need to program, you don't need to remember specific terms or anything like that. Natural language interfaces are going to have a large role in agentic AI because as humans, that's an interface that we like, we enjoy and has a much lower barrier for people to participate in.

So I think natural language and being able to, you know, type what you want an app to do, or what you want an agent to do for you and be able to go create that will absolutely have a large role in that. Again, I think it will depend on the business solution. We also know that humans are more comfortable in sort of like a personal assistant, like a co-pilot realm talking back and forth because that's how they interact with their other coworkers.

And so we really want as much as possible to have the human still work and the way that they're accustomed to working. So they might, you know, ping a coworker to ask a question. Now they might ping their, their personal assistant to ask that question.

There will be places where they'll actually go into an intelligent app because that's the best interface for them. And then they may continue to ask their personal assistant questions about that app. So they will be much quicker to learn about that app and what they're doing.

But then natural language interface is definitely gonna play a key role because of the way it lowers the barrier and allows humans to continue to interact with the technology in a way that they're most comfortable. So it sounds like what you're describing is sort of an agent first or, or assistant first, uh, approach to interacting with systems. Is that kind of what we're, we're moving toward?

I would kind of flip it around. I think it's a human first, a human led. I think that human is going to have a personal assistant like copilot that transcends their day with them, understands their productivity context, their business context, you know, how they like to communicate, how they don't like to communicate.

It's gonna be more kind of, I'll call it, connected with the human and their personality. And then I think there's gonna be a set of intelligent apps and agents mm-hmm. That dock into those places.

Agents may dock into your apps, agents may dock into your personal assistant depending on what they do. All that together we'll build kind of the new tapestry of how we work and how we move forward. But I think it's the human at the center with these technologies helping to make them more productive and giving them more time to think strategically, to be creative and to think about what they can do next.

We, we know from all kinds of studies that 80% of of people in organizations say they don't have enough time to do what they wanna do, to think about the things they wanna think. So we're thinking about how we empower that human and how they now have more time for those strategic creative things. And then this technology is, is really helping them along the way.

Tiffany, one thing you mentioned is that AI should be for everyone. And I'm curious if you could talk a little bit about how ag agentic automation can help ensure that people with disabilities aren't just included, but actively empowered as they're working and using enterprise workflows. Yeah, this is an area I feel extremely passionate about, what we've seen so far with, uh, particularly co-piloting and some of the automations that have been done in, in teams and some other places.

So, you know, there's lots of different situations that, that people with disabilities face. Um, you may have someone who has hearing loss and now with the transcript on a meeting they can fill in where something wasn't quite clear to them. You may have, uh, someone who has a DHD who focusing on the meeting and the notes.

Um, they feel like they miss out in both fronts. I think. I think that's a human experience across the board now with meeting notes and the transcription, like you can stay a hundred percent focused on the conversation, the meeting, and know the rest of that is going to be there for you.

You could flip this over to other environments like schools or education where the concept of meeting notes can help students take notes and lectures and they can have it all there. So they're focused on their learning in the moment. I meant a lot of these, uh, agentic AI pieces are gonna help humans be fully present in the moment and know all this other stuff is there for them to use later, but they're not having to multitask in the moment.

And the the numbers are showing, uh, people see the real impact to that. They feel like the quality of their work is better. They feel like they are more included, they feel like they have better performance, and they feel like the meaning of their work has actually gone up.

We're just seeing the beginning of all the impact that this is going to have for us. Tiffany, can you gimme an example where a agentic AI has provided an outsized impact above and beyond what you either might have expected or what we could have previously done? Yes.

We see many times that the spark for starting with AI is around efficiency or productivity, but what we're hearing from customers is they're seeing a number of other vectors of impact. Um, accessibility and inclusion has been a really strong one, which I'll talk about. Uh, being able to upskill and learn has been another one that's come up quite strongly.

In fact, ey uh, Ernst and Young recently did, uh, a study where they interviewed over 300 people who had been using Microsoft Co-pilot, uh, asking them how did it impact their work. All of these 300 people identified as having a disability. Mm-hmm.

And over 75% of them said they felt like copilot had made them more productive at work. They kind of laid that along three lines. One was removing barriers, 88% said they were doing better communications by using copilot than they had in the past.

They also talked about feeling more included and feeling like the quality of their work had gone up. That was over 85%. And they also talked about feeling like they were getting more meaning out of their work because of their productivity and the quality.

So that is just a tremendous, uh, like additional benefit that we're seeing from AI where organizations are able to ensure that every team member is bringing their best selves to work and doing the best role that they can. And I think we will just see more and more of this as we move forward because as co-pilot and some of the other AI continues to learn even more and more and becomes more personalized, it can even help in other ways that will be very valuable for people. Tiffany, I was wondering if you could share some examples about how ag agentic technology is being designed with accessibility in mind.

Yeah. So as you know, Microsoft's had a a long history of thinking about accessibility features in our products, whether that's been sort of an Xbox and assistive controllers or office and, and the many accessibility features we provide there. That same sort of mission is, is moving into agentic ai.

So we can think about what are the new accessibility features that maybe in the past weren't as feasible that now we can bring to the forefront. Some of them are already out. You think about teams meetings, teams, transcripts.

You think about things like co-pilot being able to ask questions across all of your graph data. As we move forward, we see even new opportunities. For example, the teams team is thinking about how today in a team's transcript you have whatever has been said verbally, you know, might be another language, might be in English, might be in multiple languages, but it's what was spoken in the future.

What they wanna do is include what was signed in the meeting into the transcript. So everybody has a complete transcript, whether that was spoken or whether that was signed. And that's just one example of the many type of agentic AI features that we feel like is now feasible that we're exploring.

So I was wondering if you could tell me about how AG automation has really streamlined very personal or sensitive, uh, processes and procedures. One of the areas that would be a, a great example of this might be human onboarding. So we each come to a new role or a a a new set of work with various, uh, backgrounds with strengths in places, things we know nothing about.

And agentic AI can really personalize helping that human on board in a way that they feel completely comfortable. They can ask many questions, they can get access to many resources, they can get recommendations and guidance that will help them learn at a much quicker pace, but not something, whereas in the past, they would've had to share very broadly with their new team that they didn't understand a concept or they didn't have this experience. Or maybe it's very difficult in a, a large conference room to to hear, uh, the, the voices.

And so Agen AI has a opportunity to really help speed up that onboarding, personalize that onboarding, and do it in a way that is really taking the human into account and helping them do that in the best way possible in a way that's sensitive to things and very positive and productive. Well, thank you very much, Tiffany, for a great conversation and, and real insight into the world of ag agent technology. Thank you, Keith.

I really enjoyed our conversation today. It's always fun to talk about the transformation that's ahead of us and how agentic AI is gonna help all of us move forward. Today, we heard a lot about agents, and I think some of the things that really resonated with me was the fact that ultimately to have success, you need to start with humans looking at processes and goals and then bring in the technology.

Now of course, there's a need for platforms that can really provide an orchestrated agent experience across intelligent apps, agents, and of course, all of the workflows that are integral to really driving real business benefits. And ultimately, the other thing that really, really sort of, uh, resonated for me is the ability of agent technology to improve the experience of people who may have disabilities, and to do it in a way that really takes into account how they're feeling and not really kind of separating them from the rest of the employee base or other customers, but to do it in a way that's empathetic and again, can really drive outcomes. Uh, my name's Karen Ker.

I lead, uh, product marketing for our secure WAN portfolio here at Cisco. And today we're gonna run through what is our enterprise networking vision and strategy and how we're executing it from a, uh, from a platform perspective. So it's two hours, but we're gonna break it up into some fast chunks of 20 minutes, uh, presentations for you.

Uh, we'll go through a little bit about our vision and strategy and then how we're applying that vision and strategy to a lot of the, uh, familiar products that, you know, Cisco builds, uh, today, uh, across our routing, switching wireless products. And then also from a management perspective, we've got a great lineup of presenters, uh, from our team across the product management and across the technical marketing teams. So I'll let them introduce themselves, uh, as they present their sections, but, uh, I just want to, you know, take a few, take a second just to thank them for, uh, taking the time today to come out here and, uh, present to you guys in person and everybody watching online.

So Cisco has been around for over 40 years. We celebrated our 40 years, uh, anniversary last year. And I, I don't need to kind of explain to everybody how important the the era of AI is, uh, to you guys, but I think what's important is we look back a little bit to some of the history here and some of the disruptions that we've seen in the market where Cisco's really been at the forefront of those innovations.

At the end of the day, we are in the business of connecting people to users and applications. We use technology every day, and whether it's in our personal lives or it's in, it's in our business lives. And so when we first started out, it was about really connecting people to the internet.

And, um, you know, we were known as a routing company and we've often been built on standards. We try to maintain open standards. And, and back then it was a lot about interconnecting different types of networks and, and getting people connected.

And then we had the mobile era where people started to move around BYOD, bring devices into the workplace, consume applications a little bit differently, and then we had the cloud era where applications started to move out of the enterprise into the cloud. And so now we're here at the AI era. All of these disruptions have forced us to rethink how we build our products and how we adapt to the changing ways that you as users consume our, uh, applications and data today.

And so what you're gonna hear about today is really how are we rethinking networking in the era of ai? So this isn't industry data. We obviously talk as, as, as a leading networking vendor.

We have a lot of customers and we talk to them a lot. Um, in the past year we've had a lot of industry customer advisory boards where we invite some of our biggest customers who are at the forefront of innovation to come in and tell us what's top of mind for them. And so what are we hearing?

And I think, uh, this three consistent themes that we, we continue to hear from our customers and, and hopefully, uh, you guys can relate to them, uh, number one is the complexity is increasing, right? I think when we talk about complexity, we're talking about the number of different types of devices that are coming into the, into the workplace, the number of different products that are being deployed. You know, in many cases you're deploying products to solve a problem at a certain point in time.

And you know, over time you've got a stack of products and they may, may or may not work together, they may be managed differently, but ultimately, how do you deal with that complexity when it seems like in this AI era, we're being asked to move faster and faster and faster. The second, uh, is around IT hiring and budget constraints. And we're all hearing about the skills gap.

It's not that there's a skills gap from an AI perspective, there's a skills gap from an IT perspective. There's a, the, the IT teams are, uh, leaders are struggling to hire people with networking skills. And then now as security starts to converge with networking, how do I find people with networking and security experience, right?

Couple that with networks that have been installed a long time ago, the person who originally installed them may not still be there. So how do I move and, and, and refresh my portfolio? And then we're all dealing with budget challenges, right?

More and more budget is being redirected to AI projects, um, or projects that add more va, uh, that are adding new business initiatives. And so we, it hiring and budget constraints is, is probably the other one that we hear about. And then AI is moving very quickly.

You know, networking has a long lifespan. You know, we we're, you know, we built some very good, reliable products that have been out in the market for a long time. Customers have, you know, typically deployed rider on a seven year timeframe.

And so how do we think about products? Um, you know, how do you make a buying decision today knowing that things may change tomorrow? And what do you, how are you gonna get some longevity out of the equipment that you're buying and installing today?

Okay, so those are some of the top three things that, that we're hearing about. So when we looked at, took a step back and, and, and looked at our strategy, how do we build an architecture for that AI ready, secure network? What are some of the key things that we should be thinking about and our product team should be thinking about and engineering leader should be thinking about, uh, as they build products for this AI era?

And we started on this journey a while back. So a lot of what you're gonna hear today is the result of some of that thinking. And there's really three key pillars that we see are, uh, core to our strategy and also are gonna be core to our customer strategy as they build out their networks for the AI era.

So the first one is, how can we simplify operations And simplifying operations isn't necessarily about having a single dashboard. It's about how can we leverage agen ops to help partner with you as IT leaders and make it easier for you to do your jobs on a day-to-day basis, right? The second one is around security.

More and more security is top of mind now, especially in this AI era where we're starting to deal with issues with deep fakes, data leakage, um, uh, and, you know, uh, bad actors coming onto the network. And more and more customers are saying, well, can I use the network as a line of defense? And so we're definitely starting to, to be asked around how can we fuse more security into the network?

And then there's also standards that are being developed, right? And NIST has developed standards around security. Post quantum is something that you may have heard of and we'll talk a little bit more.

And so, uh, security fused into a network is another core strategy for us. And then scalable devices ready for ai. You know, no longer is it possible just to build faster and faster and faster equipment.

You know, how do we now have build devices that can do networking and security at the same time, right? And provide low latency capabilities for AI workloads. So these are all, uh, these are what's top of mind for us.

I know our, our Cisco, uh, general managers are all looking at these three pillars and thinking about how do we, how do we build out our products and provide that blueprint for our customers? And the last slide I have here before we get into, uh, the platform section is really around, well, what does it mean to build AI hardware for the AI era? This is AI infrastructure, let's talk about it, right?

And you'll, some things you're gonna see coming out of, of Cisco, and we've talked a little bit about this, is, is custom silicon, right? No longer is off the shelf, silicon gonna be suitable, but what we need our products to do, uh, you know, how do we provide that high bandwidth, high performance, be post quantum ready and also provide security? Another one that's coming up is observability, right?

How do I provide deep packet inspection at the same time? So I'm really being asked to do three or four things that a networking device from the cloud era or even the mobile era is not capable of doing operating system. You know, we do have multiple operating systems at Cisco.

Uh, iOS X xe, uh, is on the enterprise side, has been the most prevalent. Uh, and how do we kind of, number one, make sure that we are making it easier for you as our customers to deploy ISXE upgrade when there's a patch without downtime. Um, get that deep observability, right?

How do you run containers within ISXE efficiently? And so programmability is also top of mind now, and, and you know, especially in this era of AI where we're starting to see APIs being used a lot more mm-hmm. To provide telemetry, right?

And data source. How is your operating system gonna be able to securely be able to communicate with other, uh, applications and, and management tools? And then from a system perspective, which is really the platform, which is how do we provide visibility, uh, programmable stay open, you know, this era of AI is more open than I've seen before, right?

Um, and there's a real conscious effort to, to really make sure that we can interop across the board. So, and then the last one, I think actually just touching on the system is power efficiency, right? Talk about it on, you know, we all think about data centers, um, being con, you know, uh, consumers of energy, but actually your network equipment does too.

Combine that with different standards. We're a global company. We have to deal with, uh, things like carbon credits and sustainability across the globe.

And so how do we provide you with more power efficient equipment as well?