How IT, Cybersecurity, and AI Shaped Business and Society in 2025 | TSG Ep. 992

Hey everyone. Nothing witty today. Merry Christmas.

Happy holidays. This is our wrap up. You're watching the gang.

Merry Christmas everyone. It's Alan Shimmel for Textron Gang. I have to apologize.

I did not bring my, my center outfit or anything else, or even Harry Hanukkah or anyone. But, um, we're here for our end of year special. We're gonna do things a little differently today here on Textron Gang.

We're gonna have one se well, not too different 'cause we'll still have three segments. One segment's going to be dedicated to it, stories for the year, and we have our great IT panel. One segment's gonna be dedicated to cybersecurity stories through the year.

And we have a killer cybersecurity panel. And then one segment's going to be dedicated to ai and that's gonna be a free for all. 'cause that's pretty much what AI is these days.

So, um, we, we haven't rehearsed this. It's kind of an improv theater that you're going to see. We hope it'll work.

But let me introduce you to our first panel, which is the IT panel. Joining us is my friend JP Morgenthal, also very good friend, Tracy Reagan and our friend Jack Gold. And joining Tracy, Jack and jp.

I have the, what we call the core four, not quite Derek Jeter, Mario Rivera at Core Four, but John Schwartz, Mike Ard, Mitch Ashley, and myself, gang members. Welcome Mike. It's 2025.

Certainly been a year and it has had its share. What do you got? Well, I think if you look back on it, you know, there's two obvious stories.

One is, you know, all things about ai, and we'll get to that in a, in a little while. But the other story seems to have been a lot of focus on cost optimization. I think a lot of people are struggling with, uh, different economies.

I mean, some people say there's a K economy out there, and it's splitting the marketplace in half. And, um, I think a lot of it people regardless, are under more pressure this year, past year, and probably going in the next year just to make things more efficient and reduce costs. But Jack, what are you seeing here?

I mean, are you catching a different vibe from it lately? No, I, I think you're, you're spot on, Mike, and I think there's a, a k curve in the, the IT industry as well right now. Um, the, uh, K curve for me is the, the cloud-based environment is still hiring.

They're still looking for people. They're still looking for specialists, the traditional it, you know, the, the, the, the guys and gals in the back room managing the server that's going down. Um, and I think you're gonna continue to see that.

Uh, some of that is cost related. Some of that is the assumption. I guess I'm getting ahead of the third panel here that I, AI is gonna take over some of those jobs.

But I think a lot of it is redirecting the, uh, budgets because number one, many companies aren't sure what's gonna happen over the next couple of years. And number two is, what kind of people are we actually going to need in our environment, in our IT groups to manage where we're going in the next two years with Gentech AI coming on board and all kinds of other stuff, uh, cybersecurity issues, there's all kinds of things happening in it. So I think we're seeing a a, a real tendency for a lot of companies to just pull back and say, you know, let's, let's be really, really cautious here where you don't know where our investments need to go.

And, uh, I think that includes getting rid of people, unfortunately. Uh, JP you seeing the same thing? What are you seeing?

Uh, unfortunately, uh, I am seeing the same thing, uh, Jack is seeing, and the, and it is very disconcerting that you have this new technology that is coming into its own and people are starting to make assumptions more than actual, uh, you know, experience with what it can do. So this, I I think part of this is that the, there are assumptions about what AI will do and what it will fulfill based upon some experimentation. Uh, but you know, it's like, um, it reminds me a little bit of the, uh, the drug industry, right?

I mean, you're making an assumption that this drug is going to perform as expected over time, but you really need five, six years of experience of seeing the impact of it before you know that it's safe for humans. Um, well, the, our industry just feeds the drugs to the humans in hopes for the best. So I do think there will be a falling out at some point, a house of cards that is kind of being built.

Um, the interesting thing is, will AI be able to rapidly be able to hold the house up and put the jacks in place before the whole house falls down? Which is it, you know, uh, different than other house of card situations, because you have this technology that can so quickly respond, uh, and mitigate, uh, mitigate issues. I mean, personally, I have seen that you can point it at a problem and it actually will resolve the issue in in minutes.

Mm-hmm. So, Mitch, let me ask you this then. I mean, we've been talking about automation in it forever, and to be honest, I know everybody's a little concerned about their jobs and their roles, but I don't know, is it better to be in it now than any other time?

Because I'm not doing so much Scot work and toil, and maybe, you know, it's just a different mindset in a different way of thinking about what it means to be in it. I think the good thing about being in it is opportunities. The greatest ones are o often caused when there's great disruption, budget disruption, ai, new technology, cloud, whatever it might be.

So automation is certainly one of it. And we're at this beginning, this transition phase into a hybrid, both procedural and also agent driven kinds of automation. So though it's very early in that space, yeah, so we're, we're, we're still gonna be living with the Pearl and the, and the Python and whatever other types, script, types of scripts that are, that are gonna be automating our processes for quite a while.

But I would think we'll increasingly see opportunity there. I think the other thing that we see saw this year is not just disruption from ai, uh, access to markets, digital sovereignty, sovereign clouds, uh, what does that mean for our business? If you're an international customer, what does it mean for a US business in terms of access to their markets, et cetera?

So that causes a lot of disruption, but there's also opportunities. We see companies introducing new sovereign technologies and sovereign cloud. So even at a macro level, whether you're inside it, I think the biggest disruption though, and I, I think the biggest fallacy was, you know, we don't need any of these people because they're all gonna be replaced by ai.

And we seem to be creating a lot of tools for those people that we're, that we don't need. So we don't, we still need them now, at least for the time being, You know, as, as much as things change, they stay the same. Mm-hmm.

There's still some basics here, right? Mitch, you know, this, I, I've spoken with you about this, right? The CIO was always told that the IT team is here to serve their customer, and the customer is the other employees of the company.

The, the function of it is both development, software development, and, and today, you know, and I think this year, but it's been coming for years, the role of software development has become an outsized piece of the entire IT puzzle, right? 'cause we're all software companies. Mark Andreessen said, even though Satya now says we're now intelligence engines.

So software development, meaning the platform we develop on, how we develop, what we develop, how we deploy, how we optimize, how we secure, how we, uh, observe feedback and reiterate, that has become, you know, the, the, the biggest chicken, the nest of it, right? By by huge. But there's still the other function of it, which is, you know, keeping the computers up and the help desk functions and, and those kinds of things that we don't think of necessarily as part of the software development job, right?

And, and I think in 2025, again, the software development piece of it, that's where the action's been in ai, right? And, and hence that's where a lot of our attention has been. Tracy, I know that's one of your favorite topics.

Well, I think what we have seen in this past year, if we really look back, is one of the largest transformations of operations that we've seen in a very, very long time. And it's moving very, very fast. Um, we are, we are having to, I mean, even, you know, even at the Linux Foundation, Linux is having to, to work to really keep up with the demand of the changes that people need to deliver new kinds of software, faster software, software that requires more processing.

And I feel like in terms of the job market, we have seen, I know many of the developers, some of the developers, I should say on our open source community have moved from being developers. They have moved to being platform engineers because they're, and you know, the more I learn about platform engineering, the more I understand that part of it is because of this massive shift that we're seeing and the types of tools that we're using, how the operating system is changing. And we now need a way to do onboarding of new employees down at the developer level as quickly as possible, and be able to implement new tools as quickly as possible.

It is, it's moving very, very fast. I mean, think about the rise in MCP servers over the last year. It's been phenomenal.

But that means that those people on the operations side have to support a lot of new developers, new platforms, new stacks. And I think that this disruption has caused some of the hiring to be quite honest. But I also think that the platform engineering has made a big, big difference in how we, how agile we are and how we develop and what we're writing.

But I don't think that AI has reached its peak in any way, shape, or form. I think that we have spent the last year trying to figure out how to build these development environments so we can start thinking about how to implement technology using this new platform that we call ai. I think there's another piece that I've seen as well, and, you know, we talked about the, the K economy for it.

Uh, I'm seeing some companies at least move parts of their IT budgets away from people and into the cloud to AWS or to Azure, you know, Google Cloud. And so what that's forcing is companies to look at it, budgets are going up, but they're not going up by that much, right? They're going up by a couple of percentage points here or there.

And so if you're spending a lot more money with Google Cloud, where is that money gonna come from? So I classify that, clarify that for a second. Do you mean managed services that are handled by those guys, or do you mean me putting my workload up there, but I still gotta manage it?

Both. Okay. Both.

It, it really is both. Uh, it there, you know, there's a reason AWS and Azure and Google Cloud are making a ton of money, right? Uh, those, and, and those budgets are not just coming from line of businesses, they're also coming from IT in, in many cases.

So that's had an effect, I believe as well. It also has an effect in the sense that companies think that if AWS is managing all these clouds, for me, why do I need all these plumbers in, you know, in-house doing the plumbing when I can get it done outside? So it is a, it's a really mixed bag as far as where budgets are, are being dispersed these days.

And the, the, the hyperscalers has certainly had an effect on the hiring. I think there's two things that are true here that Alan's point. One is, developers keep playing with new toys, and then they go build stuff and they get bored with it, and then they dump it on IT people.

That seems to be continuing to happen. And then, um, the second part of this though, is I would also argue that maybe we're looking at the democratization of it, and as we do make these advances, there's whole companies in the mid-market and smaller organizations that don't get to take advantage of all these wonderful things for years. And maybe we're gonna see these organizations implement more advanced it sooner faster.

John, what do you think? So this is taking into account what you've all said is, it's interesting you imagine being inside a company, being a CIO or being in, in charge of the IT plan or the organization where you're being hit with wave upon wave of all these options. You're trying to balance your, your budget.

You're trying to think, figure out where the money's gonna go, what's gonna happen to your employees. I think it's gonna be head spinning. I think it's gonna be a mad dash.

And I think you all see a combination of people mass hirings and maybe a lot of people being transitioned out. And I, I see that, I think about Amazon, I think about how they're using AI and how they're implementing it in within their, IT and I, you just see this kind of chaos going on. And I think Tracy had said, we're kind of just at the beginning and it's only gonna get faster, which presents a lot of problems.

On the flip side, as Mitch had mentioned, disruption's a good thing in terms of ideas, in terms of opportunity. So it's, we're kind of at that, uh, assessment process. And the assessment is overwhelming You.

The, The other thing, I think 2025 introduced is probably the greatest beginning of the greatest skill shift in our industry. Yes, every job is being not just affected by it, but the skills of people and their ability to use it. Maybe create things with it, maybe operate it, maybe all the above.

That is part of that democratization too. But you look at the skill surveys, they've, they've started to shift into new categories around managing agents or prompting and how, knowing how to, how to instruct and describe things in, in a way that AI can take on, or even looking at roles and jobs and processes and figuring out how to use and apply AI that's outside of it too, not just within it. Mm-hmm.

And Alan, the question I can't answer right now is if, you know, if your child comes home and says, you know, I want to go into it, do you tell 'em that's a good idea right now? Or not? It is funny you should mention that, right?

I've, I've had this conversation not only with my own children, but friends' childrens who come to me because they say, oh yeah, go ask Alan. He's in tech. Um, tech's not going away.

It's not going away. But the skills you're gonna need and what the jobs are, are, as Mitchell said, absolutely changing. And, um, at the end of the day, you know, this whole AI thing is just, it's going to affect every single job in terms of, it's not necessarily gonna replace or take every single job, but if you are not an effective user of ai, someone else will be, and they'll take your job.

And we've been saying that consistently throughout this year. Right. Uh, another aspect I I'll bring out, you know, all of us here are no longer 21 or 22 is, and I, I spoke about this at the, at my shimmy says last week, is the torch being passed to a new generation?

All of us have grown up with 20th century technology, right? We, we've, we've had great careers. Everyone on this show that I'm looking at has had significant accomplishments in their careers.

And we are doing our best to stay abreast. We're swimming as fast as we can, right? To keep up with this AI thing.

But is, is it, is it it, you know, the boomers are on their way out. The Gen Xs are in the back of the room already. Is it time for the millennials, the alphas, the baiters, whatever they're called, the Gen Zs?

Is it their time to take this baton and run with it and and See where this goes? Yeah, I think we already are. Yeah, yeah, yeah.

I mean, we, this iteration has happened generation to generation. I think just in this example, it's just profoundly larger and faster. It's gonna infect more people.

And, and it's, and again, going back to like the skills and, and what Mitch mentioned, that's why it's so hard to, to follow some of these surveys and, and adoption of certain AI in particular, is that we're at different stages for different skill sets and different development of skills depending on the job. And so it's really hard to get a really hard grasp on it. But is This bringing any different than when we've had transitions in the past?

Yeah, maybe it's coming faster. Maybe it's, it's more effect. Have a greater effect.

But when, when, when we move to PCs, I go, go back to many computers, right? And from mainframes, even mainframes, it always transition, transition was always hard. I Think the big issues, the big salary reset that's gonna occur with this, and I think that is gonna be the problem in the short term for many of these people that Alan's saying, we're passing the torch to their expectations have been set by us, right?

Who have had, you know, looked at for the past, uh, 15, 20 years, you know, entry level jobs in it, of, you know, 50, $60,000 because it was a skill that required a significant amount of training and expertise in order to execute. And now with ai, the, the expectation is that you don't need so much training. You don't need so much experience.

You need how to know how to use ai. But given that the volume of people who will understand how to use AI will be immense, the salary expectations of starting, I think are going to drop significantly in it. And I think we're gonna be back to 25, $30,000 starting jobs again.

And that is gonna be a problem for those who are expecting higher and they're not gonna take the job. And we're gonna see this, this, um, valley, if you will, of where there's this challenge where I want more money. Sorry, we ain't paying you more money.

Uh, and so it settles itself. It's gonna create a very interesting dynamic in the industry. Jake, don't dare you said it, But we have to stay frosty.

Don't spray what everyone else is thinking. And I, I wanna make sure that I get this in because I always like doing predictions as well. I predicted AI agents would suck.

And I'm, I'm, I'm being proven to be right. If somebody was to ask me today, should we go, should I go into it? I would say, yes, it on orbit, go learn as much as you can about satellites.

We are gonna be seeing a disruption in data centers in the next five years. We could go from, we have like 11,000 satellites above us right now today. We could go up to a hundred thousand in five years.

And what are those satellites gonna be doing? They're gonna be our data centers. They're gonna be our data centers, and they require a whole new set of skills to manage those data centers.

That's where we're headed. It's really, really obvious to me when I look at what's happening in the satellite industry. That is where we're going.

We're talking about satellites. So the size of toasters, literally, we're going to space. So technology, hang on.

'cause we're gonna be blasted into space in no time. And it's going to be a busy, busy, busy time with plenty of good salaries. It's not just new person has gone before.

Guys, we, I, I gotta pull the plug on this one. We're over 20 minutes. I got people in the green room it panel.

This is all now recorded for, uh, pros not prosperity me. And we can look at this, but we're gonna take a quick break. We're gonna switch out some panels, and we're gonna come back and look at cyber.

You're watching Techstar Gang, You've earned it. The spotlight, the responsibility, the weight of teams, companies, and entire industries fall on your shoulders, lives depend on your decisions, your home life included that work. You are protected physically and digitally.

Nothing gets through your team without a fight. But in a globally connected world, everyone sees you, including those who mean to cause you and your organization harm. And now home your sanctuary attackers see an opportunity.

Your digital front door is wide open. And what compromises your home can breach your boardroom. Because the devil's greatest trick isn't targeting your workplace firewall.

It's convincing you that your personal life isn't at risk. Black cloak, digital executive protection, defending the new attack surface your personal life. Hey everyone, thanks for coming back to us here.

So, as I said in the outset, you know, a little different with our, our Merry Christmas year end show. Our second segment today is cyber in 2025. And I couldn't think of three better cyber people to talk about it with.

Uh, we've got, and two, Santa Claes and Chris Blas and Fred Wilmont, Fred Wilmont, and the woman behind the, or in front of the brick wall, Terry Robinson. Um, of course joining Mike Mitchell. John and I thank you, my cyber experts.

Um, you know, 2025, like everything else in tech, cyber had itself a year. We saw some amazing acquisitions for big, big money. We continue to see consolidation.

What I found most interesting was after every year of hearing, where's the, the covet, you know, what a covet is? The who vet the saying, where's the innovation in security? Where's the innovation?

All of a sudden, I think we're seeing a little innovation. It's AI fueled, maybe, but innovation, nevertheless, Mike throwing it to you, It's been an interesting year, but every year in security is an interesting year. But I feel like there's been a fundamental change.

And it goes something like this in my mind. It's, um, the pace of the battle is now being fought in real time in milliseconds. And it's changing the way the security people think about their jobs, their roles, and, and maybe even the stress that goes with that.

But since we have, uh, Terry here, I'm gonna go with ladies first. But Terry, what's your thought here on, um, you know, is, is the mindset of the security team changing or has it changed? Or what does the psycho demographic look like to you?

So, yeah, I mean, I, I've had this conversation a number of times with people, you know, recently. It's like some of the issues are the same. Some of the, uh, uh, mitigations and remediations are the same.

But the problem here, or not the problem, but how, what they're having to adjust to is the speed, right? Of everything. I mean, it's, it's, uh, it's crazy.

Now we're seeing exploitations, um, of vulnerabilities, what within hours, uh, of disclosure. Um, that's gonna be minutes. Um, I think it, um, or security, it is like completely stressed over a lot of this.

They just, they already were having problems reacting, right? Or, or certainly not staying ahead of, of everything. But, um, I, I think that if you could get into the psyches of these people, you would, it would be scary.

I mean, I really do. And I, and honestly, I, I, not that I don't think that they'll adjust, I, I think they're going to maybe in 2026, I'm always hopeful about this, is that security will find a way to be more proactive. So they're not always, uh, caught reacting to what's going on in the marketplace.

But, you know, you guys can tell me if that's just a pipe dream. So Fred, we've established that cybersecurity people are scary. Um, but I guess the next question that I might come with is, Did you say that with these beautiful Santa Claus Here we, the Santa Claus hat cyber guys?

I, I needed to get a Santa Claus hat, and I would've been more rosy, I guess. But, But what are they gonna do to kind of succeed or change? Or is there a, you know, something that they should be doing?

'cause I, it feels like the old playbook just isn't gonna work. So I can sit around and stress out about everything. I need to figure out something to prioritize.

So what is it? So there's three different, uh, forces at work here. Uh, and I think I loved what Terry had to say about it, right?

The velocity and veracity of attacks weeks to days to hours, right? The, probably the biggest RC from 2025 was hours, uh, with React to Shell and the greater than 1% of the internet available for operations on that particular vulnerability. But you have a couple of other interesting and contrasting, uh, forces at work ai, for better or for worse, right?

Uh, skilled talent shortage. But also there is something that, I think there's a, I don't wanna call it a malaise or a LAC that's happening within the industry, uh, but it's within the business, which is to understand the actual risk of what's happening and to treat that risk as a business risk and cyber insurance and the things that go with it. There is no doubt at all that the number of, uh, rce this year is greater than it's ever been.

The number of exploited CEEs is greater than it's ever been. The number of technologies that have been exploited, that our security technologies is greater than it's ever been. But there are an awful lot of really talented cybersecurity people that are outta work.

There are an awful lot of folks that are struggling to implement what everybody's invested, all the financial wherewithal from the venture capital community to make work in the industry. And that hasn't been a wider gap that I've seen in the last 10 years. Good point.

Chris, what, what, what do you think you've established yourself as the, you know, cybersecurity optimist of 2025? Yeah, we gotta, we gotta understand what story we're part of, right? So lemme take you back to the winter of 1992, right?

And you find a young crisp, uh, Ebenezer Blak, uh, looking at the firewall market where there's a hundred firewalls and start on the planet that typically seven computers a million dollars a year. It was the, it was the application proxy, uh, uh, era. And I had an idea that we'd have a firewall we'd sell thousands of, but we do that, we intercept every packet.

And by nine, the winter of 1998, you find me in, uh, uh, uh, aged, uh, uh, Chris in Utah taking over the, the Cisco having just taken over the Cisco firewalls, which are not proxies. We're at this point where we've blown past that paradigm, it's not gonna work anymore. And now, you know, here we are, you know, the Christmas present back in Toronto again, in a world where, you know, our, our systems are breaking because they don't know.

There, there are people here. We see this everywhere. And security is just a good example.

Compliance, ask any CISO from the beginning of this year to this year, it's just breaking everywhere. Why? Because the systems don't know that Bob and Mrs.

Crochet are their actually trying to use the system. Literally, our systems have no idea we exist. And on every level they're breaking, right?

So the Christmas futures, you know, I think the, the moral of the story, we know pretty well our systems will come to recognize that there are people here and that changes everything. So I think that's that little moral arc. We are part of that story and everything.

You know, Terry and Fred, you just said, we're driving to this. We've hit this point where it takes a change of mind, and I think we know what it looks like. All right, Mitch, the Christmas Carol kind of opens up with Jacob Marley, and he's visiting Ebenezer Scrooge, and he's talking about how I wore these chains in life.

And so I'm now wearing them through eternity. And sometimes I think, you know, cybersecurity feels like that, and maybe Jacob Barley's the patron saint of cybersecurity, I don't know. But how do we break the cycle?

You know, it, it is a perfect metaphor, and here's why. So if we ever thought that the idea that technology could be held back until we were able to secure it before we deploy it, that has been blasted out of the universe. You know, if you wanted an example of how badly will we rush into the next technology AI without worrying about security, we'll rush.

And so it changes the entire psyche, I think for security, for security people, for the mindset, yes, you still have to be the, you know, be the folks that really are looking with great scrutiny and, you know, prove it to me. Let me make sure that this really works, and how do I protect from the bad guys? But I think the big change is, you know what, it, we are on a new cycle, and I think it's gonna continue.

Whereas it's not about wait till we decide what we wanna do, and then we'll figure out, come to you and figure out how to secure it. You better be on the AI train. You better be on the data center and satellites in, in space to Tracy's conversation in the last segment.

Um, you need to be on where technology is heading and already working, thinking about, all right, if that takes off, what are we gonna do? How, what's our story? What's our strategy?

Or maybe there's an innovation that we're gonna help create to make that happen. So the days of sitting, sitting back and saying, wait for people to come for permission to do things, if there ever really was that, maybe that was back in the 92 era there, Chris. Um, you know, the, the Christmas future is we're all flying and, and we have to move together.

Move very quickly. So when you think about, I was gonna ask you, so when you think about this, looking forward a little bit. So we have this recurring theme of cybersecurity being this, this evolutionary nature versus AI's revolutionary pace, for lack of a better terms.

Does this, and I don't wanna be a doomsdays sayer, but does this inevitably lead to a cataclysmic event that finally forces even cybersecurity to change its ways? And maybe does this lead to like a rash of acquisitions by companies hyperscalers who want to protect themselves and their customers? I'm just, it just seems inevitable this is happening can happen.

Like the cyber Nine 11, is that what we're No, maybe it's a good question though, to the point where does cyber becomes so infused into the platforms and everything else, that it kind of just becomes a feature and it's, yes, it's not its own independent category. Alan, what do you think? Ba hum.

I, I'll wait for that. Yeah, because John, right. You know, that inability is there, right?

You know, so the systems have to get back to relating to humans, right? And it sounds kind of esoteric, but, you know, and as as says, as a longtime security practitioner with quantum computing coming, I actually do see an answer to that. It's that, of course you have my information, of course you have my documents, you can't read them, you can't use them.

You can read them. They're right there. You have no idea, because that's in the relationship.

That information is in relation to Terry and something Terry is doing with Fred. How do you interpret that? So I think there are ways out of this and, but they are, you know, much more human than the bits and bytes of cybersecurity for the last couple of decades.

But they're implemented in that same, uh, infrastructure and it's seems to be showing signs of working these days. So there is hope tiny Tim. Well, and I think Hearken back to what Fred said maybe too, is, um, you know, the whole idea of, of the business proposition, right?

And, um, risk, and maybe that's, you know, what, uh, defenders or organizations, uh, focus on because there's just way too much for them to do, right? When you get down to it. And maybe, um, everybody that I talk to again talks about how we gotta know what's important to us.

It's an individual decision organization to organization, you know, what you, uh, protect. And maybe you're protecting the same thing throughout the decades, you know, just maybe writ large or whatever. But, um, you find a different way to to, to do it.

Um, I, yeah, I know. It's just, I, my my point I guess is also is that cybersecurity has gone through all these waves, so it's adapted and it's, it's yeah, for severe in many cases. And I'm just wondering though, given this kind of tsunami like wave coming with through ai, whether it's overwhelmed or It's, it'll still adapt.

I think fundamentally though, here, here's, here's the deal. Fundamentally, there is a, a schizophrenia in the goal of cybersecurity at, at, at some level. I don't know if it was Mike or, or, or John who said, you know, security goes away and becomes part of it, built into the fabric, if you will.

I think if I asked Fred that, Fred would say, that's a worthy goal. I'd love for that to happen, right? We, we need to be built in, not bolted on, we need to not be an afterthought, right?

We don't need separate security. We don't need the SEC in DevSecOps, it's just DevOps, right? That's one side of the face.

The other side of the face is none of those people give a s**t about security as much as we do. That's right. And we need, we need, we need to be here to make sure that this whole thing doesn't go to hell in a hand basket.

And, and that right there is security personified. Chris, you got your hand up. You know, in that, in that ghost of Christmas past narrative I gave you, you have to understand the mid nineties, how existential that was because the firewall was security.

And the whole concept that we're gonna accept the idea that people, some people will get through was like, apocalyptic. You can't imagine we, we'll no longer have security. And then we got into it, network awareness, SIM, sim, threat intelligence.

My point is, we've gone through these evolutions multiple times. We're at this point where driven to the point, but John, you, you know, call up. We can, so we will do something, you know, the light will stay on.

I, and you know, a lot of people tend to think that it comes down to humans, like relating to the, have the system relate to the human, because actually that gets exponentially harder to crack, easier to deal with smaller. So we're driving ourselves in an evolutionary, uh, again, to a crux where we have to do things differently. We've done this before.

I think this episode next year, we'll, we'll have some alignment and agreement on where that's going. I think it's about people. Yeah.

So Alan, to you, to your point, Terry's point about maybe a little more focusing the conversation on risk. You know, we've been talking about security, people learning how to talk to the business forever and a day now. And I think maybe some more of them are figuring that.

But at the end of the day, you talk to business people and everything they do has a risk. And when you go to business school, it's really about being taught how to manage risk. So from their perspective, security is just one more risk of many, many.

So are we finally getting to the point where maybe the security people realize that, you know, this risk issue is what the business people care about, because they just wanna know, like, yeah, I know it's insecure, but I wanna know, like maybe how insecure and how much money I'm gonna make regardless. Fred, how long have you security? Uh, it's worked 25 years.

20 plus. Have you been hearing this same argument for 25 plus years? Absolutely.

Chris, you, Oh yeah. We're all gonna die any second now. Yep.

Yeah. Same, same stuff. Different day.

I'm gonna age out of that, that prediction, by the way. We, we are, we're all still here, right? And this, and again, you know, the, the, it's interesting.

I know we're going into AI in a second, but again, these are semantic systems. We're literally talking to them and trying to see what makes sense, which we're literally doing with each other right here. And, you know, we'll either navigate into a, a Christmas future where we get to keep doing things, things or we're not.

And that binary just continues to drive the fact that we'll come up with solutions. We always do. I mean, Mike, to, to a, a good answer to your question.

That's why we have CISOs. The CISO is supposed to be the universal translator, translating security to business for the decision makers. They've gotten better at it.

CISOs have become, uh, more widespread and better understood. I also think that they are like, uh, like, uh, building department employees at the New York City building department. You slip 'em a 20 and you get anything you want, but there you go.

I don't know who insulted there. Was that the CSOs or the building department people? No, you take your pick.

It depends where, what side you coming at it from. Yeah. Boat.

That was, that was a grabber on the very unique New York thing. Yeah, that was a New York thing. But let, let, let us, let us take a break here 'cause we're over time.

We're gonna come back and do our grand panel on ai. What else should we do at Grand Panel Launch? You're watching Textron Gang, Discover Textron Group, the epicenter of tech innovation.

We are your go-to for reaching IT, leaders and practitioners worldwide. Our secret impactful content that sparks awareness, engagement, and top quality leads with us. You'll access editorial websites, streaming videos, virtual events, custom content analyst research, and more.

Join our satisfied clients. Let's revolutionize your tech journey. Contact us today and tell your story to the world in the most powerful way with Textron Group.

Hey, everyone. And we're back. You know, we saved, I don't know if it's the best, but we saved the biggest for last.

And this is, uh, our mega panel as we're calling it, talking about what else? Ai, because as I've said before, and I, I'll say it again, 2025 will go down as the year of ai, right? Where, where we, we moved into this new journey.

Um, for me, it is, it's the start of the 21st century. I think up to this time we played with a lot of 20th century technologies, the internet, cell phones, the digital revolution. I think when we look back, this will be the start of the 21st century.

And we'll have robotics or physical AI and, and, and, uh, quantum and all these things. It's gonna be a very different world 30 to 40 years from now. Much as World War I was kind of really the beginning of the 20th century, right?

That was the end of the age of empires and modern nation states and everything else that flowed from it. So as we sit here at the end of this first year of the rest of the century, Mike talk, quantify this. I look at it a little more concrete maybe.

I think if I look back at the year of 2025, it's the year that everybody figured out that the AI industry is having trouble parsing the word is in much the way Bill Clinton did. And, and the issue is they keep saying in the present tense that we can do all these wondrous things. And then you get into it and you find out, well, there's a big gap between what they say we can do with these things, and maybe someday we will, and what we can actually do with them today.

So Tracy, I mean, you're up to your eyebrows and AI stuff this past year, but what's your assessment of what's real, what's not? That's a good question, right? Well, I can say what's real for me.

Um, I have learned a lot about AI in the last, uh, 12 months. I have become far more productive, what, in terms of what I do using ai. Uh, but I still, I feel like it's, we're still pretty basic, right?

Um, sure it can create a par it can rewrite a paragraph for me and make it sound a lot better. Yes. Are we generating more code?

We are definitely generating more code, but we still have to walk through it carefully and verify almost every single line to make sure it's doing what we want it to do. And it doesn't have any bad things in it. And when I look at other countries and, and how they're starting to use ai, if we look at China, they're delivering ai.

And I, I complained about this from the, from the very beginning. AI needs to help people's lives make their lives better. We're seeing a lot of com.

The AI in the US is very focused on businesses and not so much your house. And that bothers me. I still wanna see more AI making people's lives better, more in medicine.

And I'm sure it's out there more in banking. Help me do things that I need help to do other than just write a paragraph or write some code. And maybe that's because that's how I'm seeing it.

So I think we have a lot of room to grow in ai. I think we're just learning about ai. Uh, and agents still suck, guys, and we have a lot of work to do in cyber and ai.

There is so much to do in ai. We are just scratching the surface. We, we haven't even begun really to, to do development in ai.

And I really do think that we have to find more ways to use AI to make the lives of everyday people make their lives better. It's Not, you know, a lesson I learned from my friend Brad felt in technology is that the, the companies that win, the players that win, they win because they suck less. JPI know AI has been a big, a big part of your world this year.

It Has for sure. Uh, you know, I have touched it, uh, and use the technology in many, many ways. I use it, uh, for, uh, research, for writing support.

I have done software development with it. I have done infrastructure management with it. I have built agents.

I find, you know, uh, the technology is continually surprising me with answers. Uh, the most surprising is how it, uh, it seemingly has a life of its own. It does not like to be told what to do.

Uh, much like a, uh, uh, a child, uh, emerging through its teen years, uh, I've called it a teenager multiple times on so many different w levels. It, uh, you, it tell, it comes back and tells you, I did what you asked. You check it out.

The jobs half done. So it's a teenager in that way. Uh, and it's a teenager in that, uh, it, it does, you tell it, this is what I want you to do, and it, it doesn't do what you ask.

However, if you give it some general goal guidelines, like, this is kind of what I'd like to see done. It does amazing things. And, and it will bedazzle you with its creativeness.

So, um, to me that is the, uh, epitome of AI at the moment. Uh, I, I would be very cautious about letting it run amuck and believing that, uh, it, it, its answers are, uh, pristine and, and accurate. Uh, however, uh, I think as a tool that is watched, uh, it, it has produced some incredible, uh, things that, let, let's face it, it's corpus of knowledge is immense, much more than any human could ever maintain in their head.

And on top of that, they have the ability to find patterns between the data in its corpus that most humans would miss. That is its value right now. Right?

John, do you think that the AI vendors just completely missed the boat? Because, and I'm asking the question because, you know, instead of focusing on here was how we're gonna make your job and your life better, they just went out and started banging the drum in Wall Street and said, you gonna eliminate your job? And it's like, well, if you eliminate my job, I don't have a life.

So I generally Don't. Well, they got, they, they got ahead of themselves because they saw the opportunities. They, they're looking down the road at five to 10 years, this like Mark Benioff and Jensen Wong do.

They, they think about life three years from now and how it benefits them. And first and foremost, what they did, the first thing they talked about were these digital workforces of millions of people. And my first reaction is, what does that mean to the people who have the jobs already?

And we are seeing to a small extent what it meant to some of them. It was Salesforce, several thousand people lost their jobs. Um, it's, it's interesting they got ahead of themselves because they pronounced this was gonna be the year of AI agents, which as Tracy pointed out, and I think Gina pointed out in a show last week, is a grow silver ex exaggeration.

Um, they got ahead of themselves. They didn't lay it out. They, instead of kind of giving it to us piece meal wise that they did, is they gave us a grand vision of what they think it will be and where it will be and how it will benefit them financially.

And they kind of lost their viewpoint. And I think they scared unintentionally. I don't think it was intentional, unintentionally.

They terrified a number of people and sent them scrambling and created confusion, which actually hurts them in the short term. So yes, they got ahead of themselves like they always do. Alan, is this a lie that the bubble's built on, or what?

No, the lie that the bubble's built on is that we are going to be able to spend all that money, build all these things, power them, cool them, and then charge for them at a, a rate reasonable enough to return a profit, right? com bubble was built on, right? It cost me a dollar.

I sell it for 80 cents. Don't worry, I'm gonna make it up in volume. Right?

No, you still lost 20 cents every time you sold it. You just sold a lot more. So you lost a lot more 20 senses.

It's, it's the same, it's the same thing here with ai, right? When, when the, unless something fundamentally changes the, the power consumption, the cost of doing these computations, it, it's just, that's, that's the bubble. That's the mismatch.

It's actually worse. Technology's terrible. It's great.

Sorry, go ahead. It's actually worse than that. I mean, various statistics have shown, you know, various research has shown that 80, 85, 90%, um, AI projects and companies don't produce an ROI If that's the case, and we're talking about basically AI is lipstick on a pig.

I mean, if you can't generate money for me, why would I pay you to give me those tools that aren't doing anything for me? And, and so it's, there's a fundamental disconnect between not that AI doesn't have a future, please don't get me wrong, but there's a fundamental disconnect between what people are promising to deliver today and what's actually producing for me. And if you can't fix that problem, aside from the fact that you're losing 20 cents every time I use your, your AI tool, it's, you're never gonna make any money.

Right? We're, we're Insist, and that has still wrap. I'm sorry, That has a lot to do with that has a lot to do with how we've implemented ai.

As to Alan's point, we have, uh, we need, everybody needs to take a breath step back and, and look to see how we can fix some of the problems, right? I, that's why I keep talking about small language models. We need domain expertise.

Maybe we should have a, a system that has a way to have a cons, a consensus between three expert domains, right? So the, the large language model was a good way to get it out the door, but it's not a good way to go forward. And I'm gonna continue saying that over and over and over.

And we have a problem with agents. We've gotta fix these. This is what's costing us.

And like I said in the first segment, look up to the stars because that's where we're gonna have our AI data centers to address some of the cost. You know, I don't think our problem is ai. I think our problem is we have seven, eight, whatever number tech executives all trying to be Steve Jobs all at once.

And they all suck at it. They're not Steve Jobs, which is really not An overriding Problem for sure. That's been an overriding problem.

Yeah. But even before this, yeah. And you can, you, you can have your complaints with Steve Jobs too, but a lot of those, these predictions and where they say we're going, that's about setting the vision and setting the quote unquote strategy for where they're taking their companies that's setting stock prices as well as day-to-day performance, right?

They're trying to build the roadmap of how convincing investors that they're gonna continue making them money. Um, um, And if you've bet on a limited amount of funding, you're never gonna get good product. Yeah.

Yeah. It, it's, it's a different world. It's not AI's the problem.

I think our, our, our mouthpiece are the, are are part of the I too. I would, you know, would've been Interesting to see. I would like to argue the other side of the point here though, and I'm gonna call on JP for this, right?

'cause there are people who are using AI to do amazing things, and they have a lot of skills and a lot of expertise in that space. And you could argue there might be even a, an AI divide starting to occur because there are people who have the skills to really maximize it. But for the average person, it's not quite living up on the promise.

So jp, are we on some sort of timeline curve here in terms of who's using AI in the most benefit and what we might see in the coming year? I, I think that a lot of people are, uh, overwhelmed by the information that's coming out from, uh, the vendors in the industry as a whole. I think there are people who are, have grabbed a hold of this thing and do a hell of a good job on marketing.

I if you're, if you're the type of person who watches reels on Facebook or Instagram or even LinkedIn, you know, you're led to believe, uh, that these people have created these apps that can do amazing things with ai. And so you start looking into and being able to understand what's really under the hood. I people call their, their application agent-based applications.

And then, you know, I, I wrote a, a piece on this, uh, uh, I believe it's a blog or a LinkedIn article, uh, about the AI whitewash, right? Hey, it's fine to say that your application is an AI application, that it uses AI to perform some of those functions. But calling it an, an agent-based AI when it doesn't have any comprehension of what an agent is, and that an agent is this autonomous thing, and there's no autonomy to these things that they built, it's, you know, the lack of separation causes confusion among people who are trying to understand what is this thing and what can it do for me.

So we're in that age of, you know, uh, parkers and, uh, you know, uh, used car salespeople, uh, how hawking at you, oh, buy my wares, buy my wares. Uh, eh, this is the greatest thing since sliced bread. It'll heal you.

You know, also, although mention something really quickly. Can I, I'm sorry. Tr can I just, so that's funny when you said JP, because you know, some people accuse Steve Jobs of being a PT Barnum, but I think he would've been the right person at the right time to explain to us where, why AI should be embraced by us.

And he would do a much better selling job than these other folks. Sorry, Tracy. I think he would've let some of these AI agents Out the door though.

'cause he was kind of finicky about making things actually work. Yeah. Yes, exactly.

But we Got, Tracy and Terry both had their hands up trace, go ahead. Yeah. And Terry, to JP point, we've, this is how we've always seen it.

How long did we have tools that had SQL in the name? And then we did the smart thing and we did a lot of I things for a lot for a very long time. So when we stop putting AI in the name, we know we've arrived Terry.

So, well, mine was just an actual praise the Lord from, uh, jps. Um, Oh, okay. You know, you'll, but I do have something to say about AI before we get to the end of this.

Uh, and, and agents, and I'm gonna go back to the, um, the teenager analogy. And that's where I, I feel like we are with AI agents and I mean, I'm talking about the true things. We don't, you know, uh, ask any parent who's had a teenager, we, we often don't know what, where they are and what they're doing, what they're saying and what they're activating, right?

And this comes from a little bit of a helicopter mom. So, um, that's a confession. But I think that, um, that's one of the problems that we're gonna see going forward.

How do you know what these things are doing and who they're doing with? And two, and that's, yeah, that's, we've actually, so that's my jam. We've actually gotten better at that.

The providers of the large language models have gotten much better at that. I, I, I think it was Tracy or, or Harry, I, I've, one of you said earlier about com, you know, committing to small language models, and that's something that I've been predicting for, you know, is needed for years. We need to get away.

You know, we, we, it, it's, it's gotta become a hybrid model where your business is captured in the model, uh, and it's local to you, and it's a small language model. And then when you need to get additional inputs into that model from the larger world, from the outside world, that's when you go to the large language models, uh, and you grant from the corpus. And, and we need this, you know, tiered, uh, model.

And everybody just immediately goes out to the large language models. I think it is a tiered model. I think you need to build and host your own model that understands you first, and then let that be influenced by additional data that you can then bring in on demand from the larger corpus.

I think that's the right approach to using these things. So what I heard JP say, and everybody else kind of echoing a little bit, but is this kind of like the, maybe even the Stanley Steamer era of AI and we're working on an engine next, or, you know, is this kind of where we are? Jack?

Kind of kind of wrap it up for us. Yeah, I think it is. Uh, you know, the, the notion that JP had of, and, and Tracy had as well, of building smaller models that are suited to what you need makes a whole lot of sense.

The challenge is, and I've seen this with many companies, the challenge is that I can't get my data together to build those models. I've got data, it's silos, it's scattered all over the place. I don't know who owns it.

Some of it's on pc, some of it's in the cloud. And so, yeah, it makes a lot of sense to build models that are very dedicated to what I need until we get to a point where people are able to actually gather their data together and then build a custom model for my particular needs. We're gonna be stuck with these LLMs there.

There's, that's the only thing many companies have. So there's a lot of maturing that really needs to take place in this marketplace. It's going to take years.

You know, people think AI is going to be here tomorrow. It's not, agents aren't gonna be here tomorrow. That's not gonna happen.

Good agents at least. And so I think ultimately AI is gonna be huge, but we're rushing, you know, we're rushing off a cliff if we keep moving in this direction. All right folks, I'm gonna wrap it up there.

I think if I look at 2025, it was the year we recognize the potential of ai. That's not nearly the same thing as realizing it. And so hopefully this is the worst AI that you're ever gonna see is the one you got today.

Hey everybody, thanks for sharing your thoughts, not only today, but all year long. I look forward to seeing you all in the new year, and same to everybody who's watching this show. And please stay tuned for the rest of the texture on that TV lineup.

But once again, I hope you guys had a great year. Stay safe, have a great holiday, and we'll see you in 2026.