The Compound and Friends - CrowdStrike’s George Kurtz Reveals the Biggest Cybersecurity Threat of 2026

Episode Date: January 14, 2026

On this special episode of Live from The Compound, Josh Brown is joined by George Kurtz, President, CEO, and co-founder of CrowdStrike to discuss: the state of cybersecurity in 2026, how AI is reshaping the threat landscape, how AI agents, non-human identities, and browser-based workflows expand the attack surface, common root causes of modern breaches and how defenders can reduce risk, CrowdStrike’s announced acquisitions and the move toward continuous identity and zero standing privilege. They will also cover why the browser has become a primary control plane for enterprise security, securing AI workloads, agents, and data with Falcon AI Detection & Response, the shift from AI copilots to autonomous, agentic security operations and why unifying endpoint, identity, cloud, and browser security matters in the AI era. Disclosure: Josh Brown is a long-term investor and shareholder in the CrowdStrike company. This episode is sponsored by Betterment Advisor Solutions. Learn more at http://Betterment.com/advisors Sign up for ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠The Compound Newsletter⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ and never miss out! Instagram: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://instagram.com/thecompoundnews⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ Twitter: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://twitter.com/thecompoundnews⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ LinkedIn: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://www.linkedin.com/company/the-compound-media/⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ TikTok: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://www.tiktok.com/@thecompoundnews⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ Investing involves the risk of loss. This podcast is for informational purposes only and should not be or regarded as personalized investment advice or relied upon for investment decisions. Michael Batnick and Josh Brown are employees of Ritholtz Wealth Management and may maintain positions in the securities discussed in this video. All opinions expressed by them are solely their own opinion and do not reflect the opinion of Ritholtz Wealth Management. The Compound Media, Incorporated, an affiliate of ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Ritholtz Wealth Management⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠, receives payment from various entities for advertisements in affiliated podcasts, blogs and emails. Inclusion of such advertisements does not constitute or imply endorsement, sponsorship or recommendation thereof, or any affiliation therewith, by the Content Creator or by Ritholtz Wealth Management or any of its employees. For additional advertisement disclaimers see here ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://ritholtzwealth.com/advertising-disclaimers⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠. Investments in securities involve the risk of loss. Any mention of a particular security and related performance data is not a recommendation to buy or sell that security. The information provided on this website (including any information that may be accessed through this website) is not directed at any investor or category of investors and is provided solely as general information. Obviously nothing on this channel should be considered as personalized financial advice or a solicitation to buy or sell any securities. See our disclosures here: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://ritholtzwealth.com/podcast-youtube-disclosures/⁠⁠⁠⁠ Learn more about your ad choices. Visit megaphone.fm/adchoices

Today's show is sponsored by Betterment Advisor Solutions. What growth strategy are leading RIAs using that most firms don't? Segmentation. Some clients' needs are sophisticated and required deep ongoing planning, while some clients' needs are simple, like those in the wealth accumulation stage. The smartest firms know, planning shouldn't look the same for every client, but the experience should always be exceptional. Now it can be with Betterment Advisor Solutions.

It's the platform built for segmenting your book, and streamlining those smaller and simpler accounts. The onboarding experience is automated and paperless. The portfolio management is streamlined and tax efficient. The client experience is consistent and modern. And the impact isn't just felt by your clients. It's felt across your entire practice. Imagine a back office that's humming,

a team that's thriving, and a service model ready to scale. Betterment's advisor solutions. Your biggest regret will be not doing it sooner. Learn more at betterment.com slash advisors. All right. I believe we're live. You guys, investors think about software companies in terms of TAM, total addressable market. When we think about TAM as it pertains to cybersecurity and the opportunity in the age of AI, it's impossible not to get excited. As AI workloads multiply exponentially, so does the need for cutting edge security software.

My guest today is George Kurtz, the president, CEO and co-founder of CrowdStrike, a leading provider of cloud-delivered endpoint. protection, threat intelligence, and cybersecurity services. CrowdStrike is announcing a major acquisition today, which we'll get to in just a moment. Full disclosure, I am a long-term investor and shareholder in the company. George, thank you for doing this. Welcome back to live from the compound. We're so happy to see you. Great to be here, Josh.

Always a pleasure. Awesome. Just a reminder for the audience, all standard safe harbor disclaimers apply. No one here is giving you any investment advice. refer to the 10,000 word disclosure in the description below. I want to get into the state of cybersecurity in 2026, but George, can I congratulate you first on being the best performing cybersecurity stock of 2025? Can I do that?

You certainly can. And I appreciate it. We put obviously a lot of hard work into it. A team did a great job. But when you have the right products and giving the right outcomes to customer stopping breaches and reducing their costs, success tends to follow. I think the market agrees that you guys are doing exactly that. Crowd strike is up about 35% of the last 12 months, which is double the performance of the

S&P 500, triple the performance of one of your largest competitors, which we will not name. But your stock price has done better in the last year than all of the other publicly traded cyber plays. And I guess I would just ask, like, how does it feel to be a gangster? You know, you always got to look forward to the future, right? You're only as good as your last quarter, but I think we focused on building something enduring. And, you know, when you really care about the customers and you think about building the best technology, not just getting something out there, it shows and it shows in the results. And I think part of, and we may get to this, but part of the success has been the Falcon Flex licensing where we make it just seamless for our customers to be able to buy more and procure more easily without going through another procurement cycle.

you can see the adoption just quarter over quarter of another module and another module another module. So putting it all together has really paid dividends for us. I want to put up one more chart that we custom made for this because I think it tells the story of the outperformance in fundamental terms. So let's put that chart up. George, this is your quarterly ARR history. And your ARR growth has been about 62X since 2017. I hope that matches up with the numbers that you guys keep internally. But if you are a shareholder in any software company and the ARR growth looks anything like this, you probably have made a lot of money in the stock.

And Crowdstrike shareholders like myself certainly are appreciative of that. Would you agree that that's probably among the most important reasons for the success of the stock? Absolutely. First, thanks for putting that chart up. That is a very good looking chart. And you don't see many of those charts that look that way. So we're pretty proud of that. We do track that ourselves.

And a big part of our success has been the AAR growth. And, you know, when you've been consistently above, you know, rule of 40, 40 percent at the scale that we're at, right? Yeah. It's not an easy feat. And to continue along in that direction, obviously, is something that we really work hard. But, again, having the right products with the right outcomes is important. important. But ARR is one of, an ARR growth is one of the biggest factors that we look at as well free cash flow. And that just gives you the real health of the business. Yeah, I'll send you that

chart offline. So let's start here. If you had to quantify what's different about the threat landscape going into 2026, what would you say is the biggest shift you're seeing? And what are the metrics that you guys watch internally in terms of like attack or advantages changing or methods or tactics. What are some of the things for our audience that they should be aware of that you guys are battling right now? Well, right now it's the battle of the agentics. The adversary is obviously taking advantage of agentic AI. And already. Oh, yeah, you know, for for some time now. And so are we, obviously, but it's the battle of AI. That's the way I would basically position it.

And how do we actually quantify that? Well, the easiest way to do this is a time measure, right? And over the years, you know, we've measured, you know, in weeks and days and hours and now minutes in terms of how quick the adversary can actually operate. It's, you know, slightly above 40 minutes. It used to be above 60 minutes. It used to be, you know, a couple hours, you know, a few years back. So you could see that time compressing.

And that's one of the best ways to look at it. We quoted one stat in our threat report, I think it was 51 seconds between when an adversary got onto a system and then pivoted and moved to something else. And what we're seeing is that the adversaries are having a gentic AI actually do the work. So one of the challenges maybe get into a little bit of the nitty-gritty detail is the fact that in the past, if an adversary got on in a system, they would basically have a tether. They'd have basically a line, if you will, into that system, and they would have to control it. And that was one of the ways that the security companies could look at and say, well, wait a minute, you're being controlled by some other destination. Now with a gentic AI, they can actually drop malware that isn't even malware. It's just basically prompts.

And it doesn't phone home. It basically reasons on what system it's on, where it is, what it needs to do, what data needs to gather, and it can work autonomously. So this is the big change that we're seeing is autonomous Mower, as opposed to sort of this command and control infrastructure that's taken place over the last decade. So a foreign actor or domestic, it doesn't matter really who it is, has the ability now to come into a system and leave something

and then vanish, not be tethered, not still be in the system, but that article that they've dropped into the system is now AI enabled and able to carry out whatever it's been programmed to do. It is. And it's different on each system because a lot of it is sort of scripts around prompts, right, to understand what system am I on, what unique data is on that system.

Where does that system, you know, what does it have access to? What identities are there? So it can do all of these things that an adversary would do, the best adversaries, but it can do it now autonomously. And then it continues to evolve what it does. So it makes it very difficult to say, well, this is the, way it works because each prompts gives a different outcome and it makes it very difficult to figure out that, you know, this is a piece of, you know, arguably agentic malware.

We keep hearing, quote, every company is an AI company now. So how does that change the definition of an attack surface? Is the biggest new surface AI agents and non-human identities, AI data flows, is it browser-based workflows like you guys? are cataloging all these new avenues of attack or twists on the old avenues of attack? How do you think about cyber in this age where every company is an AI company? Well, I mean, let's take a stroll back in time. You probably remember 2022-ish, I guess, you know, the chat GPT moment. Yeah.

When someone's like, wow, you know, this is amazing. First time they try it. And then most organizations look at that and go, okay, you know, back then, like, that's cool, but we're really worried about it. and then they have to figure out how to protect it and make sure that their data doesn't get put into, you know, the training models and go external and all those things. Now, in 2026, you have a lot of companies that have built their own AI agents that have access to data, they have access to workflow, they have access to compute, they have non-human identities. Essentially, they're superhumans that have full access almost to all your data and your networks and even systems outside of that. So from that standpoint, now we're in the implementation phase, early phases of leveraging agentic AI.

but it needs to be protected. So, you know, one of the acquisitions that we did last year was a company called Pangia, which really provides guardrails around implementing secure AI, if you will, and protecting AI agents. And we're doing that today because companies, like the genie's out of the bottle. You can't just say, hey, we're not going to go AI. Like everyone wants to do it. They want the benefits of it, but they have to do it in a secure way.

And it's a real challenge. So we've seen a tremendous amount of interest in things. things like Pangea, so that companies who want to save money, who want to have the benefits of AI, can actually implement it in a secure way. Right. You guys have talked a lot about stopping breaches by correlating signals across endpoints, identities, the entire cloud. In 2026, what is the most common pattern?

So when a sophisticated intrusion happens, whether it's at a corporation or a governmental institutional or whatever. What's the pattern that you see an intrusion succeeding? And then what are you guys telling people is the fastest way people trying to defend against that intrusion can reduce the risk? What is different now versus how it used to be? I mean, the simplest way I could say it is adversaries don't break in. They log in in 2026.

Think about it. And this is why identity is really the new perimeter. And when you look at a lot of these attacks, particularly from the e-crime groups, it is abusing identity, meaning they will socially engineer a help desk as an example. Human is always tends to be the weakest link. They will get admin credentials. And then they will basically, once they have, you know, access to the front door, traditional systems basically say, okay, you have access to the front door.

We don't really. You belong here. Yeah. And we're not tracking. what room you're going into or what you're doing, which is, again, you know, part of the acquisition we did, which we'll talk about. But that has been the biggest key is identifying the compromised credentials. But in addition to like just getting the password, so if socially, if someone is social engineered and they give the password, that's one thing. But what they will do is they will

actually compromise the session cookies. So you probably use, you know, a cloud email system, including Microsoft or Google or something. And you don't log in every time you look at an email. You're authenticated because you have a session token. They will steal that token. And then without logging in, they're able to replay that. And they have full access to, again, they're past the front door. So that's why you need newer technologies like Signal and some of the next gen identity technology

that we've been building out. But that is like the number one vector right now is logging in versus just breaking in. Okay. You made a $740 million acquisition revolving around continuous identity. And why don't you tell us what the name of the company is, why you bought it, and what you think this acquisition brings to the table for customers. We'll talk about shareholders after customers of Crowdstrike. Well, we're so excited about this acquisition. Actually, our largest acquisition to date.

But the technology is incredibly hot and a fantastic team came out of Google. and some other big-name companies. And they are solving the problem that I just went through, which is the traditional identity market has been sort of PAM-based, right, privilege access management, where you are at bill time, you're saying, okay, Josh is an admin or is a user, and these are all the entitlements he has,

and this is what he's allowed to do, and then you kind of forget about it. And then, you know, you leave, you move around. Who knows what happens, but you have all these entitlements. and you have all these credentials to critical systems, right? That's sort of build time. What we're talking about now is runtime identity, which is really the future, and it's non-human,

particularly agentic, meaning you have zero standing privileges. So as a user, you can do nothing until the system believes that you have something that you should be doing. So as an example, if there was a help desk ticket, if you were just an admin, you have a help desk ticket. Well, what is that ticket? what do you need to do? Well, you need to do something from an administrative perspective. It can actually figure out who you are, what you need to do, and dynamically give you the privileges

just for that workflow and just for that action. Okay. And then take them away. So by definition, you always have zero standing privileges until you get privileges for what you want to do in that workflow and then they're gone. And that will solve a lot of the problems and identity is continuous identity that has to take place that I just talked about. Okay. So for the people that don't run large enterprises or aren't in the C suite of large corporations, I'd think of this as like a household. So your daughter is graduating from college. You're going to throw a big party in the backyard. Some of it will spill into the kitchen. You have 200 guests. In this analogy, one of those guests makes his way upstairs for some reason and is in a guest bedroom or is in the bathroom or is rooting around in a closet. A system like this will stop that from happening because

that person, while yes, they've been granted access to the backyard to attend the party, they have not been granted access to be in a closet that's upstairs. And running a large organization with thousands of users, maybe millions of users in some cases, it's sort of important to know, yes, there is some access for this person, but what are they doing over there? and locking that person out of activities that just because they have access to the home doesn't mean they should be able to do all of the things that they might be doing. That's correct. And it will dynamically give you access. You could steal that, by the way, if you liked it.

Okay, I appreciate that. You always got good analogies. But if you said, hey, someone needs to go to the garage to get some more beer or something, well, you've got to give them access to the garage. It will figure that out, and then it will take that access away. The other thing that's important, specific to this is the complexity involved in this, and what happens today, because you always have to look at the outcome that you're giving to a customer, right?

It isn't just new technology. Today, literally, there are thousands of identity rules that are so complex that there's always mistakes that are made that companies can't deal with it. So what Signal allows a company to do is to actually express that in English terms to say only admins with an active ticket can make a change in this system.

So they get zero access, but the system understands if there's an active ticket, who an admin is, and the fact that they're working in the company, not internationally or what have you, and it can express a rule. One rule can collapse 10,000 rules that are prone to mistakes. This is the power of what they built. They built an identity graph. So they're really an identity fabric in this architecture, and they connect all these disparate systems and rules to be able to make it so much easier to implement.

Is the business risk here that this could force corporations to completely overhaul the way that they are classifying, not only their employees with real identities, but the tasks that those employees are meant to be involved with, plus all the non-human identities that are now being introduced into the system like is part of the challenge here, okay, great, now we're way more secure, but companies have to be thinking about the way that these new systems are protecting them and how they have to self-organized to make it work. Because I can envision a nightmare scenario where an IT department at a company is getting calls like, why am I locked out of this? Why am I locked out of this? I can't get my work done.

I know that there's a ton of thought that goes into that problem, but it sounds like that's what this acquisition is meant to address. Correct. And when you look at, like, we haven't closed the acquisition. We're still going through that. We announced the intent to acquire them. So that'll happen, you know, over the next however long. But when you look at what signal is doing, typically they'll get into a pilot area. So a large company, and they have some really big customers. We'll say, we have a problem in this area. And, you know, we're not talking about enterprise deployment. We want to solve this problem. And they solve it. And they're like, wow, this worked and it just was seamless. And then it begins to, it begins to expand within that. And we've seen that in looking at the company and working with the company across many large enterprises. So that's typically the way we would see the rollout. Now, again, there are a smaller company.

We've already rolled out with our identity technology. We're already on the domain controls. We already have all the signals that come together. This is the power of the model and the data model we have. We have trillions of signals that will go into signal the company and help basically provide better identity decisions. And so we feel really confident about it. But to your point, there's some re-engineering that takes place.

But the great thing is companies, they're, you know, I always say you can't push spaghetti. You got to pull it. Like, they want to go there. They're leading us there. And we're going to have the solutions with the Falcon platform to deliver on it. In the press release for the deal, it's mentioned that the browser remains a critical cybersecurity blind spot. 85% of the workday is spent there.

Existing security models, either force users into Waldgarten enterprise browsers or rely on high latency network routing. So I have 84, I think 84 employees right now, let's say. If you told me they spent 85% of their day in a browser looking at some system or another, I'd believe you. I'd think that sounds about accurate. But why is it so critical to separate the browser from the security or to make sure that

these two things are working in concert with each other? Yeah. So let's just make sure we segue, which is this is the second acquisition, seraphic, different than signal that we announced today. Okay. And I think part of it, I'll answer your question here in a minute, but part of it too is, well, we spent over a billion dollars in acquisitions that we announced in the span of a week. Why did we do this?

And we did this very deliberately, which we take acquisition and shareholder capital allocation, incredibly serious, which is identity is the new perimeter, but the browser is the front door. And if you think about your employees and expand that to the largest enterprises, the bulk of what they do in their day is in their browser. They're in an email. They're talking to Salesforce. They're in workday or service now or something. you know, it's all there. Like, you don't even need a VPN in many cases because everything is sort of SaaS enabled and browser enabled. So why do these acquisitions complement each other? What's the strategy behind this? Well, seraphic and signal. Like, how do they work together?

Exactly. You've got continuous authentication. You've got identity, which is one of the biggest problems. And now you're combining it with the visibility within the browser. So one, you need visibility what's happening in the browser. And then two, you need the controls to be able to implement, well, you should only be logging into the system or you should only be uploading this data or downloading this data or you want to watermark a page so that somebody can't just copy it out of there. So you have the ability to provide incredible visibility and control combined with our agent signals to allow companies to implement control models that they could only dream of. And that's for their employees. The second piece of that is all these unmanaged devices.

As you might imagine, there's a lot of companies that have third parties they work with. Could be an outsourcer. It could be some help desk or third parties where they can't control the entire operating system. But what they can do is they can give them a secure browser that only allows them to do certain levels of access. And that really replaces things like VDI. So this is one we're incredibly excited about. There's a whole range of technologies that we can replace with this. and it provides, again, visibility protection control,

not only in security, but in data protection, and preventing insiders from doing bad things. And they do it in a way that doesn't require you to get a new browser. A lot of their competitors say, Josh, you know, we can do that, but you've got to throw away the browser you use and use our browser. This one runs underneath all the browsers, and we can implement it seamlessly.

Bring your own browser as big. I want to pivot to Falcon. In December, you announced the general availability, ability of Falcon AI detection and response. Wanted to ask how that's going so far. What are you hearing from customers? And why was that a meaningful announcement or development? Well, it's meaningful because there's pent up demand.

Right now, there isn't a company of size that isn't looking at how to protect AI, period, around the globe. So we had announced it in the fall. Pangaea was the technology, the company that we acquired, Great team, great technology. We generally made it generally available mid-December, and literally within an hour we had our first order. And it came through Falcon Flex, which means, again, part of our licensing is, like, you don't

have to go through another procurement. If you want something and it's on the price list, boom, you can just pull it down, and then it begins to burn off. So it's one that just had a lot of pent-up demand because people have budget. They have a mandate from the board. They have compliance needs. They have to solve this. from just from a compliance perspective,

and we think this is going to be a real winner for us. So obviously, it's early in the life cycle of these technologies, but we spent two months sort of going through integrating it. Again, part of our goal as a company is when we buy these things, we don't just buy them and throw them over the fence. We make them work seamlessly within the platform. You guys announced a cybersecurity startup accelerator with a cohort of 35 different startups,

and your partners on that are, AWS, Amazon, and Invidia. So I have that right? You have that right. Okay. What are the gaps in the ecosystem that you think these startups are going to help to fill? Or what is the exciting part of that sort of activity from a crowd strike shareholder perspective? Well, I love doing this.

I'm personally involved in it. And in security, there's plenty of gaps. You heard me talk about the technology curve. And it's sort of up into the right over the last 30 years. while security has to parallel the slope of that curve. And there's 6,000 security companies that solve very bespoke security needs to larger, you know, sort of market opportunity. So there isn't one company, crouch, or, you know, pick another company, any of the big public companies. We can't solve everything.

We want to solve a large part of it. But what we see in these startups is they are in the early days of solving something that's unique. And it could be, you know, a big emerging market or it could be very, very important. bespoke to things like IoT, but it gives us a way to get involved with these companies early. We actually have something called the Falcon Fund. We invest in these companies. In fact, Xerafic was an investment we made over a year ago. So we actually get to help, nurture, see what's happening. So we got to be in the ecosystem early on to make sure we understand who the future winners are and what customers like. And then, again, being very thoughtful about our M&A strategy,

it doesn't mean we're going to acquire every company in this thing, right? Because they're going to go often do great things as well, but it does give us a leg up on what they're doing and how they're solving future problems, and we want to be part of that. I want to end by just asking you about the AI opportunity in general for CrowdStrike and for cybersecurity. People hear things like AI helps the attackers, and then they'll hear AI helps the defenders. And obviously, as we start hearing about this year's crop of breaches and cybersecurity incidents, we're probably going to simultaneously hear the term AI in every one of those stories, which I think lends urgency to the crowd strike investment case, my personal opinion.

I would assume that you do as well. How should we be thinking about all of the amazing things that companies will do with AI this year and the role that you guys will play in making this possible? Yeah, well, one of the things, and I've talked about this, I had a big keynote last year when we continued to talk about it and deliver on it, is sort of the agentic security operation center, which is the SOC, okay? And the whole goal for us, and I mapped this out and made a similar sort of mapping to autonomous driving.

There's five levels of autonomy and driving, and there's five levels of autonomous security. There really isn't a fully autonomous car. per se, and there isn't a fully autonomous sock. And part of our goal is to deliver on the autonomous sock and security AGI, if you will, where it's basically making decisions for you and implementing security at, you know, lightning speed. Now, why is that important? When we think about the advantage that we have at Crowdstrike,

for the last 10 years, we have something called Falcon Complete, which is our managed detection response. We've been stopping breaches, and we've been, basically, we've got a treasure trove of security data and how attacks work and how we actually prevented those. Think of that as the Reddit of security. Right. You've seen it all so far.

We've seen it all, right? You know, you have different manifestations of what we've seen it. Well, guess what? And I think you probably would agree. If you look at a lot of the frontier models now, they're being commoditized, the real moats are the data. Okay? So you've got 10 years of data, which is actually annotated.

and, you know, it's reinforced, you know, human learning that we have. So we've been able to pair that together and get incredible outcomes. I mean, talking about four days of work for an analyst to figure something out, and our system can do it within an hour. It's incredible. And you're searching your own case studies, like your own library of breaches and horrible things that have happened that you guys have been called in on to help fix. Well, that's doing the training.

Like, that isn't just like, hey, what did we see? This is like, this is the way the attackers work. And then we've built security reasoning models, bespoke models that we've built based upon how we do it. So there's a lot, the key here, Josh, is there's a lot of IP that we've built that creates moats. And when people say they have AI security, like you may have AI something, but having the data is a different story. And we know we've got a massive moat and an opportunity for the future for our customers and shareholders. So I think that's really a key point. You guys, have you guys have trained models based on real-life incidents that you have the de facto single

source of truth on what happened, how was it fixed, what were the detection moments. You're not training a model, Googling news articles about breaches. You guys were fundamentally involved. I love that. It's a great place to leave it. Is there anything else that you think is worth saying on the topic of cybersecurity in 2026? Did we about cover the main points? We covered a lot. I mean, I think it's going to be a very dynamic 2026. And you're going to see the speed of these attacks increase and the speed that an attacker, that a defender has, or that time is really going to be compressed. So, you know, it's AI-led, it's helping companies do what they do best in enabling them to be successful and really building the best single-agent

platform in the market that isn't Frankenstein together. What big part of our strategy has been acquisition, but big part has been organic growth. But anything we acquire, we make sure it works together for the best outcome for our customers. Well, George, you guys at CrowdStrike crushed it for myself and fellow shareholders last year. Just wanted to say congratulations on the new acquisitions. And wishing you all the best this year, let's stay in touch and have you back.

And guys, make sure you are following CrowdStrike. thought leadership in the space, their blogs, et cetera. They're putting out tons of great stuff as well. And we'll talk to you soon. Thank you.