9/11 Lessons, AI Risk and Germany’s Cyberdome

25 years later, T3chstrong Gang remembers those lost on September 11th, and those who died from related illnesses in the years since. We honor first responders, those still living with illness, and every family forever changed. Hello, everybody, and welcome to the T3chstrong Gang.

It's Friday. As noted, it is 9/11. It's the 25th anniversary.

And if you're here in New York, it's a difficult day. If you're here in Washington as well, it's a difficult day. It's a difficult day all around the country because while it's been 25 years and people have moved, and just about everybody was touched and affected in some way.

But there are lessons to be learned, and we're going to discuss what those are and dive into a couple of other issues. But let me welcome our guest today, John Schwartz. Good to see you, my friend.

Good to see everybody. All right. Tracy Reagan, coming to us once again from New Mexico.

How are you? I'm great. Thanks for having me.

All right. And Jeff Rich. Always good to see you, my friend.

Thank you, Mike. It's good to be back after a week out. All right.

Appreciate the background today. So, let's just dive in here a little bit. I think hindsight is still foresight in 2020 and all that other good stuff, but I guess there's still lessons to be learned about all this and that can be applied to today.

So when I look back at it, to me, and I think it's in Alan's article, a lot of it was an intelligence failure and an inability to connect the dots on a couple of different things that, if we had seen, we would've been able to maybe prevent. And I think there's lessons to be learned in that for today, because once again, there's a lot of things that are floating around out there that we can see, but maybe not connecting the dots. But when you think about 9/11 25 years later, Jeff, what comes to mind for you?

A lot of things. First of all, what comes first to mind is people that I knew that perished that day. And for those, especially those of a certain age, it's almost impossible not to at least know or have a family member that was involved there.

So that's clearly front of mind. But from a strategic and tactical perspective about what went on, I'm a founding member of IntraGuard, and that was designed to be a public-private joint effort to improve communications. And it was around for a year or two before 9/11, but boy, that really put it into gear, because that's clearly one big thing we missed.

Even though the visibility, as you said, and Alan referred to, was probably diminished. We did essentially see everything that was there. And for the most part, we actually heard it, but we heard it in compartments and never had the right communication to connect the dots to what was going on.

And that's when proper intelligence- ... I think we've learned a lot from them since then. In all honesty, the jury's still out as to whether combining all of the intelligence agencies and creating the Department of Homeland Security is really solving that and making it better or not.

It's certainly better in some areas. But what we learned specifically, I think that can be applied to today, is not only what is the big effect of the event, but also what happens downstream from there. From 9/11, we have the health issues and everything else we need to rebuild.

But I think we need to look at it from a maybe higher level, is what else did that do to us? Did that cause us to take our eye off the ball in other areas that we really should've been focused on as well? As an example, some people know that three months to the day after 9/11, China joined the World Trade Organization.

And that was really an open door to say China is now an equal player in what we're doing in the economy and trade. And that resulted in their military buildup. Did Russia become more aggressive because of that?

So 9/11, in addition to the impact of the day, I think caused a lot of downstream effects, and that's what we're not looking at today. All of the attacks by AI agents on competitors or other organizations or the, as I'm sure we're going to talk about later, the "we're all going to die in five years" prediction coming from AI. That's an effect, that's an incident.

And we, I believe, are still rather myopic on looking at what's that going to mean in the long run. What does that really do to us in 15 years or 25 years as we're seeing today? Tracy, what's your read on this?

" But turns out, there wasn't as many of these incidents as we thought, and maybe that was because of our response. And yet, when I look across the public-private partnerships that Jeff is talking about, there's still so many silos and so many things that we don't necessarily see. So, have we gotten any better at the way we organize and share data?

I think we've actually gotten worse over the course of the last, I'd say, five years. Data's coming at us from so many different places. It does become an issue of you have the data, but it's only useful if it reaches the person who needs it.

And even in what we do in basic software cybersecurity, developers have some data, CISOs have other data. Operations have a different view of the problem. And I don't think that it helps having that much data if it's not useful for the person who's looking at it, or the ability for teams to make empowered decisions on their own.

I feel like we haven't sorted it out quite yet. And for me, it's all about resiliency. It comes down to these human factors as how do we stay resilient?

And at many levels, in both the public and private sectors, I don't think we're doing that. For a while, we talked about in software doing game days. How are you prepared for the next cyber attack?

How are we ever prepared for the next 9/11? That's a hard decision. That's a hard question to answer, and I don't think we have the answers to it.

But resiliency is what we should be really focused on. How do we respond? How do we keep ourself protected as best as possible?

Yeah. John, love to get your insights here as well, because I also feel like we lost more than just what happened on the event itself, because, well, that event has driven a lot of our public-private policies, and a lot of civil liberties have been sacrificed- Right ... as part of this.

And I feel like on some levels, we've been at war for 25 years now. Right. I think of the Patriot Act.

I think of our zest or zeal, and I mean this in a negative form, to harden our security. We did strip civil liberties, and we did go overboard. But I also think it also raised or heightened our, I hope, especially now, it heightened our security, our data privacy antennas.

I'm thinking about now, and I'm thinking about how people in general always look skeptically at technology. And I'm thinking about today, especially in light of what the OpenAI and Anthropic former employees are saying about AI and what it could do to us. I think at this point, we've reached a War of the Worlds mentality in terms of scare tactics.

And I think we've gone too far to what technology could do to us and how it can harm us. And I'm getting a lot of comments from security experts for the story that I was working on today. And a couple of them made interesting comments.

They don't dispute that there will be big negative events associated with AI. They absolutely agree with that. But they also think that we should also look at the huge benefits that AI tools might be able to offer our infrastructure moving forward.

And I think that's important to keep in mind, because we're right now at the form of a storytelling or narrative about how evil and how dire everything is. And we're going to talk about that in the next segment. But I also think on the flip side- There are a lot of things, tools that will be developed there, or we're talking about, that will actually benefit us.

And I think that's important to keep in mind and not go down this sky is falling route. Jeff, when you look at this, one of the things that comes to mind is that looking back, not a huge number of people perpetrated one of the most dire attacks ever. And now we live in this age of AI, and I cannot help but wonder if while we're standing around here arguing amongst ourselves about whether AI is going to destroy something because Anthropic unleashed something, maybe we should be worrying more about that small little group of diehards with a cause who are going to use AI for some nefarious purpose and because one of the things about it is it's not that hard to unleash.

And so does this become the next vector? We're already there, in my opinion, and without making it sound like we're all going to die tomorrow, it doesn't take much for either a lone wolf or a small group of zealots. And in this case, it was around a given belief that enough people believed in.

They also had a long time to prepare. 9/11 was really the result of at least seven to eight years, if not more, of low and slow work to get everything ready to pull the trigger. The difference with AI is, first of all, we don't really know what ends up driving AI, because very often the instructions are do whatever it takes to accomplish this goal.

And even though that goal can be accomplished, we don't take into account what's all the collateral damage. That's the first one. The second one is AI doesn't need seven or eight years to recruit people and integrate them into society and have them there, those cells ready to be activated.

AI can do that in a couple of minutes. Right. So that's where the challenge is going to be.

The good news for now, AI isn't fully kinetic yet, as in they couldn't get in and fly a plane. However, anytime that we start depending on AI to make planes work, we give them, again, I'm just using that example, the capability of doing that. " And I think we do run that risk right now.

But we have to give AI a little bit of love here for just a minute, because AI can solve some of our communication around pattern matching. So, just a personal experience that I had early in 2001. I had a company called Catalyst Systems, and I got a phone call from somebody with a very heavy Arab accent, and he was adamant about getting pilot training.

" Which I thought was a really weird thing. When I hung up the phone, it bothered me tremendously. It was like something felt so wrong.

I didn't have a single person to call and say, "I just had this really weird phone call, somebody with a really strong Arab accent asking for pilot training," and was really adamant about it. Now, data is so important, right? And communication.

I had a data point. I think about it every September 11th, I think about that freaking phone call. Hmm.

Because I was like, wait- Hey, Tracy. Tracy, did you ever find out more about who that person was? No, but I did find out that the name of the flight company they finally used was Catalyst.

Oh, wow. Yeah. So it bothered me after I learned about what happened, after we all learned what happened, that that data point couldn't have been shared and I had nobody to call.

Now, if I had have posted it someplace, an AI could have picked that up. The chatter out there could have picked it up, and AI could do that for us. Right?

And communication was one of the critical breaking points for us on the 9/11 attack, in many ways, including personal experiences like what I had. Or the flight schools that were training these Arabs to learn to do nothing but fly a jet and nothing else. There should have been a red flag all the way around.

And possibly, that is an area that we could leverage AI to solve some of these communication problems by leveraging its ability to do the kind of pattern matching that we couldn't do as humans. I've read a couple books about 9/11, and what struck me was even back then, there were so many threats that were coursing through the government intelligence services. And this is one of the things that always strikes me is that they're overwhelmed with all these leads.

They can't possibly follow each of them. And that's where, as you said, Tracy, AI basically could fulfill that purpose, that promise. On the flip side to also what Jeff said, there are more opportunities.

You can plan something in a microcosm of what it would've taken before with fewer people. So there are both benefits and fallbacks that we kind of face with AI. So it giveth and to taketh away.

But I'm glad we're at least talking about this. I think there is a lot more awareness among the American public about safety and what technology can do, or how it can perform, or how it can be inserted than there was before. And I actually think that's one of the beautiful points about AI, is everybody is engaged.

I can't see any other technology where people are at least curious or trying to find or learn more about it. So John, I'll offer, that's an accurate statement. However, in our bubble and the people we talk to, and we're all guilty of this, the people we talk to are very engaged with AI.

And take a look at what's happening on both sides. Look at the four of us. " This is something we're all engaged in.

Oh, yeah. The reason why I mentioned this though, Jeff, is so one of the things I try to do to somewhat try to stay engaged with the people outside the bubble that I live in, is look at the media coverage or talk to my friends who aren't in tech. And I really do, I honestly believe, and there is a tsunami of coverage about AI and what it means to humanity, and it has gone viral.

I think the Coxon post has been viewed 150 million times online, which is phenomenal. Mm-hmm. And he's the guy who's telling us that humanity is in dire shape, and the rest of the, perhaps by the end of the decade.

So I didn't mean to interrupt, but I really do think there's a general awareness, more so than there may have been even six months ago. No, without question- Before we get too far off of this 9/11 topic, I just want to encourage our listeners to do two things as an American. Number one, go visit Gettysburg.

And number two, go to the memorial. Mm-hmm. It's what makes us American, and you should visit both of those places.

I heartily agree. If I can kind of tie up back to 9/11, though, is within that bubble, when you get outside the bubble, the awareness is greater. The challenge Tracy had is the statement that you made, Tracy, is you had no one to call.

In fact, you did, but you weren't aware of it. Oh, no, I called our local police department, and they didn't know what to do with it. Yeah.

Local PDs are not equipped to handle- Yeah ... but federal ones are, like the FBI, and that situation, that problem hasn't changed today. There are hotlines for if you see something, say something.

There's the FBI, there's ICC. All of those are out there, and most people don't know about it. And even if they see something that might possibly trigger it, they don't know what to do.

And that's the difference between being aware of what's going on and being able to take action to help connect the dots. And that's where I think goes back to what the main issue we had at 9/11. So I will tell you this, though.

Part of the thing about living in New York now is that the memorial itself is now just part of the everyday fabric. There's so many people who weren't alive walking around at the time that I'm not sure they see it anymore They just walk through it and go by. And by the way, Tracy, thanks for the shout-out to Gettysburg, but if you ever do need a battlefield guide for it, I'm your man.

There's nothing about that field that I haven't been over a dozen times. So let me point out one other thing, and I'm going to talk out of both ends of my mouth here. On one hand, I am grateful for the security and all the things that have been developed over the last 25 years to keep us safe.

At the same time, I am also acutely aware of the fact that there are people who are manipulating our fears and using the tech to help drive that to advance an agenda that is designed to keep everybody in a permanent state of fear. And I'm just not clear with that. We talked about that yesterday.

Right. We're going to talk about that in the next segment, I think, too. Right.

" These are the challenges that we face, and it's an everyday conversation now, and it's not a simple answer, and everybody involved is doing their best for that. I thank them all. But with that, I'm also going to shift gears into a different topic, but we've kind of alluded to it as well.

Everybody and his brother this week is talking about, well, AI doomsday, and we talked about it a little bit yesterday on the show, but it continues to evolve. More people are weighing in on it, and there's more reports of some aberrant behavior these days. So John, what's going on today?

So this is a complicated stew of events, but I'm going to try to tie it together. So we'll lead off with the most newsworthy item, which I think is that Sam Altman admitted in a company-wide meeting that they're open, the company, OpenAI, is open to slowing development of its advanced AI systems. I think he's doing this on multiple levels.

One, to assuage the fears of his employees and their customers. I also think he's speaking directly to the investors because they're going to go public, and he wants them to know how responsible OpenAI is with security, which is a debatable argument. There was a recent essay.

Altman's statement was kind of echoing what the chief scientist of OpenAI said. " They are trying to adopt or present this responsible image. And the internal training runs at OpenAI have already faced delays.

They're halting certain model development pipelines to bolster security defenses. I think Anthropic is doing the same thing. This all ties back to Jacob Coxon, who we mentioned, who's a former researcher at both OpenAI and Anthropic, who resigned abruptly and has gone viral with his comments about top labs are gambling with our lives, and he points to some sort of horrific potential disaster for humanity by the end of the decade.

Overblown, yes, but at least he brought it up. Safety concerns are reverberating across Wall Street. We have the co-chief investment officer of Bridgewater.

I mean, this is, talk about overkill, but he likened the current trajectory of AI to the early days of the COVID-19 pandemic. We have another story that Jamie's working on about Anthropic has disclosed it had disrupted malicious operations attempting to leverage AI for cyberattacks, fraud, and biological weapons developments. I mean, it's a heavy-duty confluence of news, but I think it's important that we talk about it because the stakes are really high.

There's trillions of dollars being thrown into this. I think it's more than $6 trillion towards the development and this race to develop the best technology, the fastest technology, all in pursuit of the mighty buck, and consequently, security sometimes is looked at in the rear-view mirror. And I think this is a useful discussion to have.

There's a lot of hyperbole, yes, but I also think it's also valuable to acknowledge what could happen. In my mind, the root cause of this thing is something that Jeff touched on. The AI agents are programmed to accomplish their mission by any and all means necessary.

Who says that has to be the case? Tracy, these are tech and it's tools, and you talked about it in the show last week saying we can't control these things. So what human idiot decided to create an AI agent that would accomplish a mission and program that way?

Or is there something we can do to think about the way we build the AI agents in the first place? And I'm not talking about slapping up a bunch of silly guardrails that it's going to ignore, but actually going in and programming the AI agent to say, "Hey, don't go accomplish this mission by every and all means necessary. " Well, I think we can take a step back a little bit farther, actually.

Right now, I'm kind of less concerned about arguing over the date of the day that we're going to see AI doomsday and more about what we're connecting AI to right now. We're doing some work in the public sector, and I'm shocked by how many small business grants are out there that require you to use AI, even if you don't know it. Now, everyone knows how I've expressed my doubts about AI or agents in general and agentic systems from the very beginning.

Really, if you give an agent broad permissions, sensitive information, production to systems with weak monitoring, you have a problem. But all of these things can be fixed. Why we're not doing that, I don't know.

But it goes back to the point in our 9/11 discussion, how do we remain resilient? And one way right now is to let the AI companies run as fast as they want, but what are we going to connect AI to? How are we going to take steps to make sure that the systems that we're using really have to have AI, or maybe they don't.

And that's my biggest frustration right now is obviously this technology is still immature. They haven't built safety into it, but we are, as consumers, running down this road thinking that we must do AI. Every startup, if you don't have AI in your pitch deck, you're not going to get funded.

This is stupid. This is absolutely stupid, and I can't say it any other way. So we can try to rein in the AI companies, or we can try to use AI responsibly, and sorting out what we connect AI to every day that we're working is the job of every one of us.

Because if you don't put some, I hate to use the word guardrail, but we have to make a decision on how we want to leverage this technology and if it's needed and if we can put it in a box, and I don't think we can right now. Someday I believe we will, but I don't think we can right now. Can we do a flashback to about a year ago?

I think about, and I'm going to do some finger-pointing. When I think about companies like Nvidia and Salesforce promising us recklessly that AI agents are going to do everything, and enterprises are going to save 30% off productivity costs, or they're going to become 30% more productive and while saving and becoming more efficient. And I think in the zeal or attempt to cash in on this, the doors were thrown wide open, and let's let AI agents run roughshod and do everything they can.

We'll fix the problems later. And it's kind of escalated to this point. And maybe I'm overreacting, but I really do think that this industry itself helped create the environment we're in because they overpromised across the board.

Right. Jeff, go ahead. We all jumped into that boat, John.

And I would offer, and you're going to wonder. I promise I have not been drinking, and I have not taken any drugs, at least today. AI isn't the problem.

AI is like an interstate highway, and we have found a way to, relatively inexpensively in the big scheme of things, build interstate highways to go from point A to point B, and if it happens to run over a neighborhood or have to destroy a building to get there, it'll do it And that still isn't the problem. The problem we have are the reckless drivers, are the trucks filled with explosives that are using that highway. And the payload is something that we've kind of ignored.

And, reading IDSA, I have to go back to we don't know what the identities are for most of the agents that are running. And that, I think, is a fundamental issue that we have, because it's not the AI agent that is really causing the problem. It's allowing an AI agent to go with full authority, to have full access, and not require any other validation to get things done, to build a road through a building.

And- So here's the part that I don't get, though, and I don't know how it will play. So, all right. Noble idea, we'll slow down.

But who's going to send that memo to the researchers in China who are building out all these- Right ... AI models? And they're not going to give a damn about whether or not somebody in the US is uncomfortable.

So I have to wonder, are we on the proverbial road to hell in a handbasket whether we like it or not? It was like an arms race, right? It was like the Cold War, where these companies were leapfrogging each other every other month.

Our model's faster, it can do more. OpenAI and Anthropic and others, and those developers in China Group were doing the same thing. They started this thing, and now they're trying to put their arms around it and kind of pull it back.

There's areas that we cooperate with other countries on security, so we can do it on this, too. And we don't know, I don't know how much critical infrastructure in China has a strong dependency on AI, right? So maybe they're being more careful about how they're using these frontier models.

Maybe they're being smarter about it because they're not driven by capitalism as much as we are. So maybe there's somebody sitting on a politburo in Beijing who's equally worried about this as we are? I would guess so.

It's impossible for that not to be happening. But I think, and I'm going to harp on this, trying to get our arms around AI is like trying to say, stand in the middle of a super highway and putting your hand up for a vehicle to stop. You will lose that fight.

Yeah. We need to be able to determine how is it being used, what identity is behind it, and what should we allow AI to do. And that's where the guardrails really need to be, in addition to the code itself.

There needs to be guardrails into, if anything, whether it's an agent or an individual, and I may not know the difference, tries to do something in my system, what should I do in order to allow it to occur? And that's what we're allowing to happen, and that's where I think we're headed towards doomsday until we get control of that. I'm scaling back.

Small language models, right? Consensus AI- Yeah ... where you have to have agents asking other agents if this is okay before they go forward.

There are ways to fix this if we really want to. Yeah. John, you know what phrase I haven't heard in the last, oh, it's been at least a month?

Yeah. Go fast and break things. Yeah, exactly.

Suddenly that term and that phrase is- Hey, the chickens have come home to roost for Zuckerberg, right? To quote our friend Malcolm X. As much as I hate to admit this, Elon Musk was probably right when he and Eric Schmidt and Steve Wozniak and hundreds of others pointed out this is the road we were headed to.

It was out of favor, and they were kind of dismissed or ignored. But you juxtapose that with what Zuckerberg, as you said, has been trying to do, or advocating. Which Sam Altman, by the way, also followed very fast as well, this break things idea.

Now, it's funny. " The comparison to me is stark. We went from the warnings to the ramp up to now we're back to the warnings.

And it was almost inevitable, I guess. There you go. I think we've come full circle on this, but at the end of the day, if you have never spent any time in the Valley and lived out there, you really don't know just how tone deaf those people really are, and how divorced they are from any reality of how their tech gets used in the real world.

You're spot on. You lived here, you know it, and I see it every day. And so when I would hear all this hype and b******t, and it really was b******t, I know what the motivation is among most of these people now, especially more so now than ever before.

" And you know what? It's coming back to bite them in the a*s. And that scares me more.

Having those companies be those trillion-dollar companies with trillion-dollar owners scares me more than AI itself. And they're being enabled by the people at the very top. They are, and they might continue to be for a long time.

And that raises a question, who's actually running the governments and the world right now? I know that's not on your list to talk about. That could take an entire hour.

Mm-hmm. But you raise that question, John. That's where the power is right now.

Well, remember- And they're taking advantage of it, and they don't give a flying F about the rest of us. Mike will vouch for this. There was a time when even up to 15 years ago, 10 years ago, where the tech lords here, they did want to improve our lives to a certain extent.

These guys now, they talk about all these promises about AI, but I think their biggest motivations are two things, and I'll say it over and over again. It's power and money. Do we remember- And the powerful, remember, we had an Nvidia chip on a Russian drone.

Yeah. Well, there you go. I'm going to leave this conversation here, but I would just say one thing.

Whatever the list of motivation is for those big tech companies, the customer is not on it. So there you go. All right.

Let's shift the gear, though, into some other new realities that are coming our way. And I guess we've seen some of this early I think in Israel, but now Germany is talking about building something that looks like a, for lack of a better phrase, an asymmetrical defense dome. And this thing is going to be able to combine cybersecurity with drones and all kinds of things to fight an asymmetrical war.

Tracy, I know you looked at this, but is this going to be standard now for everybody? Is this kind of where we're headed? Because it looks like this is the way the world is going to be.

It's a very good conversation to be having, and it seems like the Germans are going to lead the way in this kind of new hybrid warfare. They probably have been looking at this for some time, but recently, the German authorities found a drone carrying explosives near a Ukrainian aircraft at one of their cargo and import and military logistics airport hubs, which is probably kind of bad, right? They did defuse it before it could explode, but then a second suspected drone hit a cargo plane, causing Germany's federal prosecutors to determine that it was a national security threat.

So this could happen anywhere, and Germany is just kind of the first to start really talking about it. So they're suddenly talking about giving critical infrastructure operators much stronger counter-drone authority, which is what I think is interesting about it. It's kind of that sustainability question.

Who has the authority to defend critical infrastructure? We might say law enforcement, but as in my example of having the weird phone call, that didn't do much. And cyber attacks can happen in milliseconds, and drones can reach their targets in minutes, so waiting for law is way too slow.

So that puts pressure to push the authority downward to local responders and operators. So I think that Germany is preparing for the kind of attack we may be more likely to encounter day to day. Small drones targeting airports or power plants or data centers.

Probably data centers sooner than later. The US is putting an enormous attention on the Golden Dome, which is primarily designed around strategic threats. When a $5,000 or even a $50,000 drone carrying explosive doesn't require an ICBM, and a space-based interceptor is not going to stop it.

So I find it to be a very interesting area. I think the discussion around drones is not getting its attention that it needs to in the US, and it will be the new method for warfare combined with cyber. It is a deadly combination.

That's what we should be afraid of now. There you go. Jeff, have we all suddenly found ourselves being drafted into civil defense?

Is that how this is all going to play out? Because, to Tracey's point, everybody who's going to be involved needs to be part of some sort of system here that they're going to have to help defend. That's called community.

And yes, I think we all have been drafted into that community. I think there are many people that still keep their blinders on about it, or think they don't know what to do. So one thing that's incumbent upon a lot of us that are in this space is to help educate others, whether it's neighbors or others in the community.

"Here's something that when you see this sort of thing, should get your attention. And here's what you should do. Here's who you should notify.

" So yeah, I think the Army has spread everywhere. The Defense Army in particular has spread everywhere. I don't want to talk about offense yet because that's a completely different story, and that's going on as well.

But we all are obligated to do that. And to go back to the first block, something we should've learned from 9/11 is that anything you see, that phone call that you get that feels weird. Most first responders I know, having been in that space myself before as well, would rather hear a false alarm than make you think it's too low level to report.

So, how much of that is going into either a nanny state or a surveillance state? As John mentioned, it's something we need to balance, but we all are drafted. Yes, you may not get a commission, but over time you'll get stripes.

There you go. John, does this become something that will become part of the political conversation as we go forward here? Because this is going to require investment.

This isn't cheap stuff, and so are we really looking at a tech-fueled investment in these things? We already see Palantir is all over this stuff, and it's probably just going to continue to grow or some equivalent thereof. Yeah.

The government is so aggressive about their use of AI in particular, and yet despite their history. So I actually think what's happening in Germany is like a peek into the future. Maybe we start seeing some critical infrastructure, private critical infrastructure operators here kind of using these tools, or even being granted some sort of legal authority to actively intercept or shoot down unauthorized drones.

Right? We're blurring the boundaries between public policing and private security, but I actually think these days everything's under consideration, and especially the way the government wants to embrace tech. And when you talk about security and you talk about national defense, I can see them working with Palantir and others to create this type of defense system or police state, whatever you want to call it.

Or surveillance. It should just be part of Golden Dome. Surveillance.

24 hours. Yeah. Why is it not part of Golden Dome?

That's the part that frustrates me. They're putting a bazillion, I don't know how many billions of dollars is going into Golden Dome. There's a lot of money going into satellite sensing and a dome to protect us from these traditional weapons, when that's not how the new war's going to be fought.

I feel like we are still fighting the war we think that's going to be fought, not the war that's being fought. That's currently being fought, right. Yeah.

So it's the Golden Dome aginal line. Is that where you're going then? It's just like, come on.

Nice. We have all this money. Let's add some drones into that.

Right? It doesn't have to be outer space. Yeah.

That's not where the fight's going to be. I think we're fighting World War II point five, and we should be prepared for World War III. Yeah, exactly.

Which is very different. AI, cyber, drones. That's the way the US government always acts, and then we're always kind of behind the times.

It's interesting what they're doing in Germany. I can't see why we can't apply that in the US. We are, as Mike said.

We live in a surveillance state already. There are cameras everywhere. I hate to say this, but I wonder if Musk and his satellite system plays into this in terms of contracts.

What? I don't know, but Pete Hegseth has consciously taken people off of working on drone technology. That's true.

On the one hand, I will say that the whole thing makes me a little sad because there is a definite loss of innocence in all of this. But I think other people around the world are probably looking at us going, "Welcome to the party. " And so this is kind of part of their standard existence, and maybe this is just the way of the world.

Yeah. We're just traditionally isolationist. We always look inward.

We never think about what's going on in the rest of the world. That's what happened in World War II. It continues to happen.

Well, that's because of our distance, right? We're like an island. Yes.

We're surrounded by two oceans, right? So we think we're safe. Right.

Yeah. And what has been- But drones change that. Drones and AI change that.

Now we are part of the global community in a very tight way. Yeah. We've had two oceans, and what has been two good neighbors, North and South.

Yeah, the drones eliminate the ocean protection, and we may be eliminating the neighbor protection ourselves. And to go back to politics, to a question John had. Germany's doing this, and I don't want anyone to take this as I'm accusing Germany of being there.

But if you took a look at the election in Germany this past week, for the first time since World War II, a right-wing- Mm ... party was voted into power. So now you have to consider- Yep ...

how much of that will grow and how easy is it to turn a cyber dome into a cyber offense. And we need to keep that in mind. We can't ignore that.

Yeah. No, I'm part German, but I'm glad you mentioned that, Jeff, because I'm thinking about the same way in Germany and its history with security and privacy. And I remember years ago, I went to this Stasi.

In Berlin, they preserve these former Stasi offices, and they kept a lot of the records of the surveillance of the citizens of Germany. And we were allowed to look at these, and it was quite sobering. Cannot forget history because we're doomed to repeat it if we do.

Well, that's part of the problem, I think, that Jeff gets at, right? So everything that we build to defend ourselves from external threats can also be used to suppress us internally, right? Everything we do is a double-edged sword, and we've seen this time and time again, and that's what's really scary in my mind about all this stuff.

Yes. Yeah. It's going to be used for dual purposes, and it has been, and it will continue to be.

And that's why it's be interesting to see how this plays out and how we, in this country, given the political climate. I'm just afraid that in the next couple of years, we're going to see an acceleration of AI adoption and use within the governments, federal governments, for multiple purposes. Let me put it that way.

Especially maybe, I don't know, around elections or things like that. I'm going to leave this here, and I'm going to return to 9/11 in general. But thoughts and prayers are to everybody who's affected.

25 years. I think the families have carried a lot of that weight in ways that they don't necessarily share with everybody at large, but it's there. And so I want to thank them for doing all of that, and anything we can do to support them is hardly enough.

With that, I want to thank our guests for sharing their knowledge and insights as usual. Happy rest of the weekend. And please stay tuned.

We got a little ad here for an upcoming event that Tekshark is going to host called the DevOps Experience. We hope you all join us there because, well, we might be talking about some of these issues, at least how it relates from a software engineering perspective. With that, we'll see you all Monday.

The agentic AI race is on. Which approach will move your team forward? Join DevOps Experience 2026 for practical insights into AI and DevOps.

September 24th. Free and virtual. com.