Black Hat USA 2025 | Jake Seid, Ballistic Ventures

Clips
More from Black Hat USA 2025

Jake Seid

Co-Founder & General Partner

Ballistic Ventures

play_circle_outline Introduction of Jackie McGuire and Jake Seid at Black Hat 2025.

play_circle_outline Jackie McGuire introduces Jake Seid at Black Hat 2025, discussing cybersecurity ventures.

play_circle_outline From MIT to Venture Capital: Jake Seid on the Essential Evolution of Cybersecurity in Our Daily Lives

play_circle_outline Importance of cybersecurity evolving to an existential necessity in daily life.

play_circle_outline Securing AI Investments: A Case Study of Noma Security’s Startup Potential and Market Architecture Evaluation

play_circle_outline Market state for cybersecurity investments, highlighting recent IPOs and acquisitions.

play_circle_outline Emerging Trends in IPOs and Private Equity: High Valuations and M&A Opportunities for Innovative Cybersecurity Firms

Info
Transcript

Jake Seid, Ballistic Ventures

Jake Seid

Co-Founder & General Partner Ballistic Ventures

Join Jackie McGuire of theCUBE Research as they engage in an insightful conversation with Jake Seid of Ballistic Ventures at Black Hat 2025 in Las Vegas. This discussion delves into the evolution of the cybersecurity landscape and the pivotal role of artificial intelligence in enhancing security measures.

Jake Seid discusses their journey from engineering at MIT to building significant companies in cybersecurity. They highlight transformative changes in the industry, such as cybersecurity becoming crucial for daily activities and the emergence of AI ... Read more

explore Keep Exploring

What is the context of the discussion at Black Hat 2025 involving Jackie McGuire and Jake Seid? add

What is Jake Seid's background and what led him to his role at Ballistic Ventures? add

What is your background and what interests you about Ballistic? add

What significant experiences influenced the speaker's transition into venture capitalism focused on cybersecurity, and what motivates their mission in this field? add

What are the key factors to consider when evaluating an investment in a company that secures AI? add

What are some indicators of the current state of venture investing and acquisitions in the cybersecurity sector? add

What are some recent trends and significant events in the cybersecurity investment landscape? add

bolt Powered by CUBE AI

Jake Seid, Ballistic Ventures

search

Jackie McGuire

>> Welcome back, CUBE community. We are at Black Hat 2025 at the Mandalay Bay in Las Vegas, Nevada. This is Jackie McGuire. I'm your practice lead and principal analyst for Security, your favorite security media host. And I am exceptionally excited to be joined by Jake Seid. Jake, thank you for joining us.

Jake Seid

>> Thanks for having me. - All right. Jake is a general

Jackie McGuire

>> partner at Ballistic Ventures. But we were just talking about your background, and I think you have done... People usually tell me, "Is there anything you haven't done? " And I think you're one of those people too. So you very casually started with, "Well, when I got out of MITs..."

>> A long, long time ago.

Jake Seid

>> Yeah. Let's not admit how old we are. So take us

Jackie McGuire

>> through your background and then what led you to Ballistic, and what you're interested in right now in 10 words or less, if possible.

Jake Seid

>> In 10 words or less. So I started as an engineer at MIT. I had a chance to build the fastest zero to a billion business in Cisco's history in the late '90s. A very special experience, and then I became part of the team that launched Lightspeed Venture Partners in 2000, and spent over 11 years there. And then what I saw in cyber was a fundamental change. It went from being nice to have to being existential to our daily life. You can't get in your car, go on the plane, you can't get the operation that you need if cybersecurity doesn't work, and that's an important problem to solve. So not only can you build big companies as a venture capitalist, but you could build important companies, and so you could do good and do well, and I thought that would be a great mission to be on. And so we got a great team together to form Ballistic Ventures.

Jackie McGuire

>> That's amazing. So it's a really wild time to be alive right now. And I would imagine as a venture investor, it's an even wilder time to be alive. I was asking you before we started, there's a lot of noise in the market right now, right? I'm watching companies who've never had anything to do with AI rebrand to being AI companies. And we're seeing this fundamental tectonic shift in just the things that security teams are responsible for, the things they're expected to know, the things that their customers within the enterprise are doing. So with all of the stories, when you're talking to companies and meeting companies, how are you silently judging them in your head? So how are you thinking about, "All right, there's a fundamental difference between this and all the other companies I've talked to. " What are those key things that really pique your interest?

Jake Seid

>> Yeah. I think Noma Security is a great case study in that. So Noma Security secures AI, and there are a lot of companies that secure AI. That's one of the areas that's very noisy. And before I made the investment in Noma, I looked at the whole landscape, and a couple of things really stood out. One is teams, right? You look at founders and you're trying to evaluate founders on their vision, their ability to execute, their ability to bring talent together. And so, of course, that was something that really stood out. Also, it really is what is the architecture based around? Is it an architecture that can be scalable, that can be applicable to a number of different aspects of how you secure AI, or is the architecture going to cause the company to be much more focused in a smaller part of the market? And Noma's approach allowed it to be very broad. Now, as a startup, you have to have a beachhead ultimately.

Jackie McGuire

>> Yeah. - But it's a platform that has a beachhead,

Jake Seid

>> but ultimately they could extend over time. So that was a key way of how I thought they could really break through. And we announced a hundred million dollars round last week.

Jackie McGuire

>> That's amazing. Congratulations.

Jake Seid

>> So we've got it validated.

Jackie McGuire

>> That's exciting. So do you tend to like companies that you think can expand horizontally versus vertically? Do you tend to prefer companies who have, obviously they have a beachhead, but versus companies that go really, really deep into one specific segment?

Jake Seid

>> It's a great question. And I think whether you expand horizontally or vertically, what we're really looking for is, is the market big enough for you to be a highly valuable large public company?

>> Right. - And so that's something that can be done

Jake Seid

>> through both approaches, kind of verticalization

Jackie McGuire

>> or being more horizontal across other areas. But of course, the strategy, the approach has to make sense for the problem you're solving. But really the underlying question is, does your architecture allow you to expand in a way that you can be a valuable public company?

>> Yeah. And so as you mentioned,

Jackie McGuire

>> securing AI is one big emerging segment. There's also AI for security-

>> That's right. - ...

Jackie McGuire

>> which is a much different thing.

Jake Seid

>> Very different. - Are you seeing companies in that space

Jackie McGuire

>> that are doing interesting things?

Jake Seid

>> Is that somewhere that you're looking or is it separate from what you're doing?

Jake Seid

>> Yeah. No, that's absolutely interesting. And obviously everybody's thinking about how AI agents can help us be more secure. And if we look across every control point in an enterprise, we think about agents can make that work actually better. And ultimately, the reality is humans cannot defend against AI. AI will have to defend against AI. So for better or worse, we're on that track where we need AI to defend ourselves.

Jackie McGuire

>> Yeah. It's an unfortunate reality, I think. So as we're doing that though, I think one of my biggest focuses is the infrastructure that we need to support that AI. So are you tracking that as well? Because one of the things I'm very closely following is actual physical supply chain of components, everything from generators to chillers, to even the labor to build these data centers. Is that something that you're... I'm just wondering if other people are thinking about this because from my perspective, how fast... If we're going to deliver all of the capacity needed for all of the value baked into the market, we're going to have to build data centers at an insane clip. Is that something that you're concerned about at all?

Jake Seid

>> I think there is a challenge when it comes to energy data center capacity. But going back to the infrastructure element, the area that we focused on actually is something that's been overlooked in security, which is runtime. How do you secure runtime? And traditionally, security has had a very difficult chance to get into runtime because people who run the business, they don't want security slowing it down or creating stability risk. And so this is where we said, "Hey, if we can actually have runtime security without the trade- offs, that's going to be the new way that this infrastructure has to be secured. " And so companies like Oligo in container and application security, so CADR is a category that's emerged, or companies like Ambit that's runtime identity, identity access management for non-human and agentic identities. This element of thinking about the infrastructure from really a runtime aspect is how we've been playing it.

Jackie McGuire

>> Yeah. I think non-human identities is another area that I'm really, really interested in because we're looking at agents that could potentially do somewhat human things. Instead of deterministic outcomes, they're making probabilistic outcomes-

>> That's right. - ...

Jackie McGuire

>> which is both exciting and scary

Jake Seid

>> because I think it's easier when a human messes up because there's somebody to blame, somebody to fire.

>> Yeah. - So it's bad

Jackie McGuire

>> because a lot of times companies just fire one person

Jake Seid

>> and don't fix the processes that are the problem, but with AI, there's not really anybody to blame. So from my perspective, I think identity is probably one of the verticals within security with the largest TAM growth over the next five to 10 years just because that's an exponent. So are you thinking similarly?

Jake Seid

>> Absolutely. And one of the things, we've made three identity bets companies like Veza, Next- Gen IGA, Ambit, Next-Gen IAM, and off mine, which is identity protection. And all of those fundamentally connect to agentic identities, non-human identities. Because here's the thing, it turns out there's going to be more non-humans than humans. It turns out they replicate faster. And so what happens when a company has a hundred x more or a thousand X more non- human identities than human identities, well, then that becomes a driver. And I'd share one perspective we have, which is you talk about agentic. Agentic right now almost looks like a version of RPA where they can make choices, but they're choices that fall within a flowchart. Ultimately, the next step, which will happen very quickly, is your AI employee. And you really think about an employee is super non- deterministic, and this employee will have an email address, and a phone number, and a Slack handle, and they will be treated like an employee and be able to do anything that employee can do. That's going to be a whole new level of non-determinism that has to be secured. And so very quickly going from workloads to agentic, to an AI employee, I think is the thing we think about in terms of where security needs to go.

Jackie McGuire

>> Yeah. And I think if you think about those two AI employees start talking to each other and where if an AI is interacting with a human, there's going to be some amount of time and delay for things to happen. When two agents start, it's instantaneous, right?

>> That's right. - So how do you think about the circuit

Jackie McGuire

>> breakers and guardrails you need to let agents talk

Jake Seid

>> to each other without them getting together and going off the rails and dropping sequel tables?

Jake Seid

>> Yeah. Well, this is why security becomes so important where, one, companies like Noma that secure your business agents, your first party agents, you have to understand what those agents are doing, what's normal, what's not normal. This is where identity comes in from an authentication and authorization perspective. Is this non-human identity or this agent authorized to access this database? Is it authorized to make changes to the database? And so really we have a certain number of control points that all need to be rethought to make agents secure but not inhibit them. So that's the trade-off, right? Is no business is going to accept the agent's productivity from being constrained. So security companies that do well will make you more secure and enable business.

Jackie McGuire

>> Yeah. Funny you mentioned that. We just interviewed somebody earlier talking about living off the land techniques. And the thing they were lamenting is the same tools that admins need to do to really do their jobs are the same tools that attackers are using to really screw people. And so I think there's a point there that you can't really... You want people to be able to experiment. You want people to be able to try new things. It's just how do you put those guardrails in place to make sure that... I think understanding intentionality is another thing. The intention of the agent, what it's doing beyond just the actions it's executing. Trying to understand where it's going is another thing I'm seeing companies dig into, and trying to assess not just the action, but the intention behind it.

Jake Seid

>> And it's increasingly difficult because the agents are getting increasingly sophisticated, not in a linear way, but in a nonlinear way. And so Claude Opus from Anthropic is now able to write the next version of Claude.

>> Of Claude, yeah. - And so when agents start upgrading

Jake Seid

>> themselves and making themselves more intelligent,

Jackie McGuire

>> and start moving faster, it goes back to why you need AI to secure AI because now just the humans aren't fast enough. And oh, by the way, imagine the attackers. They're not using guardrailed AI. They're using unchained AI to do their work. So even more reasons why we need AI to secure AI.

Jackie McGuire

>> Yeah. All right. The last part I want to talk about is the markets in general, just the IPO market, private equity venture capital. And what do you think the state of the markets is? So I also host Enterprise Security Weekly, and we cover fundings, and I think we've maybe had half a dozen IPOs since the pandemic in the security space. It's just been a very, very difficult market it seems. And I think part of it is that, but the other part is that companies don't really need to IPO as much anymore.

>> That's right. - They seem willing to raise nine

Jackie McGuire

>> figure nouns-

Jake Seid

>> That's right. - ...

Jackie McGuire

>> before they IPO. So as an investor,

Jake Seid

>> obviously your whole job is to get your money back, hopefully-

>> Hopefully. - ...

Jackie McGuire

>> at some exponent.

Jake Seid

>> And so with the IPO market, not off the table, but not as healthy as it was pre-pandemic, how do you think about that, about what the next rounds look like, how your investments are going to exit?

Jake Seid

>> I think we're really in a golden age for cybersecurity outcomes and venture investing. You think about the biggest acquisition ever in the history of venture capital across any segment of venture capital is a cybersecurity company, and that's Wiz. You look at the recent exit of CyberArk at 25 billion. In my prior portfolio prior to Ballistic, I'm an investor in Cato.

>> Okay. - They just raised a nine figure round

Jake Seid

>> of 5 billion valuation.

Jackie McGuire

>> And so what we're seeing now is the ability for cyber companies to raise money at public market type valuations, get acquired at public market type valuations. And oh, by the way, what was one of the biggest IPOs in the last 18 months it was Rubrik, which was a cyber focused company. And so whether it's IPO, whether it's MNA, whether it's a late stage financing, there are so many options now to really build incredible value in cyber. I view it as a golden age.

Jackie McGuire

>> Yeah. I mean, I think there's some work to be done on the part of the exchanges to make it easier to IPO. I think IPO on ramp was one of the things that pre-pandemic that happened that was supposed to make it easier to ramp up to SOX compliance and all of that stuff. But there are lots of other options now, and I think from an... I know a lot of CEOs where IPO was like their dream because that's what you were told in the 2000s and 2010s, right?

>> Yeah. - Is that every

Jackie McGuire

>> brilliant IPO that's what you're supposed to do.

Jake Seid

>> Now you have options. Now you have real options.

Jackie McGuire

>> Now you have options that don't involve having to comply

Jake Seid

>> with a bunch of regulations.

Jake Seid

>> Yeah. It turns out being a public company CEO is a lot of work.

Jackie McGuire

>> Yeah. All right. Amazing to have you. Thank you so much for joining us. This was a great conversation. I would love to have you back because I think it's really important to bring this kind of perspective to security. Because I don't think that we necessarily always inherently think about the markets and capital, and IPOs and things like that in security because it's just not our domain expertise, but it is important as we're thinking about how we build products, and how we need to build products that continue to be funded because that's how we continue to fix the problems in the world.

Jackie McGuire

>> That's right. - So as much as we all want

Jackie McGuire

>> to be altruistic

Jackie McGuire

>> and just save the world, we need to build profitable products to do that. So I think your perspective is really great for founders who are building cool things and really would like to do something cool with them. So thank you.

Jackie McGuire

>> Appreciate it. This is Jackie McGuire for the CUBE.

Jackie McGuire

>> We are at Black Hat 2025 at Mandalay Bay in Las Vegas.

Jackie McGuire

>> We will be back with great guests, a lot

Jackie McGuire

>> of amazing information, so we hope you'll stick with us

Jackie McGuire

>> and tune back in.