Cyber Defense Radio - Jeremy London - Keeper Security - Hotseat Podcast - 2026

This is Cyber Defense Radio with your host and cybersecurity champion, Annabelle Thomas. Annabelle brings to you another globally recognized cybersecurity executive in the hot seat today. I'm your host, Annabelle Thomas, and joining me today is Jeremy London, the Director of Engineering, AI, and Threat Analytics at Keeper Security, a company that's focused on zero trust, zero knowledge, identity security, and privilege access management. Jeremy, welcome to the hot seat.

Hey, thanks for having me, Annabelle. Yeah, super excited for you to be on today. Now, I know identity has become one of the most complex challenges in cybersecurity today. It's no longer just about managing employee access, right?

We're seeing a lot of organizations are dealing with thousands, sometimes even millions of identities across users, applications, services, systems. What are you seeing right now as the biggest challenges for organizations, and what are they experiencing when it comes to identity? Yeah, identity has definitely changed over the last few years. AI and AI agents have kind of taken over the enterprise space.

So there's citizen developers building their own tools, building their new workflows out. So that has opened up kind of this new era of identity management. used to just be your username and password and logging in on one laptop. And now that same user might have 10 to 20 different types of identities that they have to manage.

So we've noticed that a lot of enterprises struggle with dealing with people building tools, leveraging AI, trying to build out new platforms, and they want to connect them to their Google accounts and databases. And all of a sudden now this one single point of contact has turned into a big mess for IT and security teams to try to manage. Right. And I know Keeper does it differently.

Keeper is very proactive at the forefront of anything identity related. I know your platform is built around that identity for security, right? You're talking about password management, secrets management, privilege access, zero trust, all of that. Can you talk a little bit about Keeper's platform and approach as it comes to identity and what sets you apart?

Yeah, totally. So the biggest component that we try to focus on is unification. One of the hard problems here is you can have all different solutions and have them kind of be disparate and discreet and try to connect and talk to each other. But it makes it very difficult to know where the password come from, who has access to this password, can this other tool access what it needs to, and can an admin kind of audit the trail of user activity through that.

So that's been the number one piece that we're trying to set ourselves apart with. We have a single plane that we kind of look through. So your passwords are in the same locations that you might access the systems from. Everything's interconnected, and all of the same enforcement logic exists around that.

So if you don't want users to have access to certain passwords, they're not shared automatically. You have to go through and kind of set up your applications and set up your groups and roles to access that. From a password perspective, that's kind of the core business that we've started in. So we have our secrets vault or password management vault.

That's where all your records and user groups and everything live. And then recently we've been moving towards privileged access management where you can now take those same user records and actually authenticate sessions, other sessions, the SSH, RDP. And then beyond that, there's SDKs and CLIs where other tools can integrate back into that. So if I need to get a password for a database I no longer storing that in a plain text file I just grabbing it from the password vault using those CLIs or SDKs We seen that process really grow as AI agents have been tapping into those systems Now when an AI agent needs to access something, they can call into those same tool chains now and be able to access those systems.

Right. And that's a huge thing with a lot of the AI adoption that's happening across all organizations. is usually people adopt AI and they don't think about the identity side. They don't think about the security side until they've got a huge mess on their hands.

So that's always something that's awesome to see that you're incorporating. Now, a major theme that I feel like is across security is least privilege and just-in-time access. Can you talk a little bit about that? Why is standing privilege such a major risk in today's environments, especially with everything that's going on with adopting all of these new tools.

How does that kind of affect the identity side and the security side for organizations? Yeah, you're hitting right on the biggest point of the era is standing privilege has worked for many years. It's still been risky. A lot of teams just do it because it's simpler, it's easier.

The next generation is looking more at ephemeral accounts, meaning we create them on the fly. Those ephemeral accounts have limited permissions. They have least privilege. And as soon as that session's over, that credential's gone.

You can't reuse it. You don't need to reuse it because if you need it again, we'll make another one. So just-in-time access is a big component to that. Maybe we don't want to be creating admin accounts at 2 a.

m. Maybe we do. And having some controls and a workflow around that allows teams to kind of move more proactively rather than just having an admin account with everything open. And as we've seen AI agents in the news be able to pick up those admin credentials and possibly do things that we don't want them to be doing, or moving down a path that a traditional user wouldn't have gone down and potentially doing something that we didn't want it to be doing.

So all of that least privilege concept is really important. Being able to create fresh accounts for those systems that need them and then be able to rotate those accounts out when they're not needed anymore really reduces a lot of those threat vectors of just having a standing privilege available. One of the big problems that we see with Windows admin controls is that's the number one route that a hacker is going to be looking for. If you get that, that's the golden key.

You can get into everything. You can change passwords. You can make new accounts. So instead of having that persisted, you can take that away, take that threat factor out, but still have a means to kind of elevate yourself using an ephemeral account to go still do the jobs that you need.

Right. And that's huge because that, like you mentioned, eliminates a lot of the risk that organizations are seeing across the board with that. Now, I know you mentioned earlier, Keeper really, really focuses on everything being in that single pane of glass. And Keeper's always introducing new capabilities like the secure database access, right, eliminating those shared credentials and untracked access paths.

Can you talk a little bit about why are systems like databases still such a high-risk area for organizations, and how does centralizing and controlling that access change the game for organizations? Yeah, so many of the latest attacks, we've seen them for years, info stealers go after hard credentials, things like database passwords, API keys, secrets and tokens, with the main goal of compromising the database, sometimes to steal information, sometimes just to encrypt it and hold it ransom.

So putting a stop in that attack vector is a very important feature You really wanna try to guard and safeguard those databases especially when customer data is there Sometimes companies have compliance and regulation affecting them So a breach could be affecting their bottom line. It could be affecting customer trust. And eventually that would lead to a company's demise. So hackers love that target point.

It's a high profit area for them. If they get in, they can potentially extract things. hash passwords, user information, and then go back and sell that. So it's a very lucrative process for them.

And we've seen a lot of the supply chain attacks this year target InfoStealer-type tools, where that's their main goal is get keys off of the system and into their system so that they can then continue attacking you or get into the databases, dump them, delete them. So it's a really important piece to monitor. Our database tools try to take away the credential implementation. So we auto-inject the credentials, we create those ephemeral accounts, and users are then scope-limited based on that new resource.

So only credentials that are generated by us are going to be able to be used and not those standing credentials, standing user privileges. So as an attacker makes a foothold anywhere, the second they disconnect, that user is useless again. So it makes it very difficult for them. They're going to get tripped up in that process.

And every time that they're connecting or moving through that system, a lot of audit logs are being generated. So we can easily build alerts and kind of inform the admin, there's something fishy going on right now. Right. Now, I know one of the other things that Kiba recently introduced is the quantum resistant encryption with the, you know, using the NIST approved algorithms.

That's not something that most companies I feel like are even thinking about yet. So I'm excited to ask you and learn more about it. But why is it important to start preparing for the quantum era threats now and even taking steps into future proofing identity security? What does that really look like?

Yeah. So our security model is very secure. We take cryptography as kind of a core concept for a lot of our product lines. And what that looks like is storing everything in a zero-knowledge way.

So in our database, it's just a bunch of encrypted data. Now, a hacker in the post-quantum space can potentially have gotten that data and then decrypted it later when the compute power makes it a simple process. So post-quantum encryption is a really important concept for us because we want to be at that cutting edge, make sure if there is a breach, if there is a problem, not even a quantum computer could be used against it. From a customer standpoint, it doesn't look any different to them.

They're still storing their passwords, but the way that that password gets encrypted and then stored safely really encourages a hacker to not look at us. It makes it very difficult to use any data if they were to try to break in and get that. From a compute perspective, it's challenging because post-quantum encryption that NIST is looking at takes a lot more effort, takes a little bit more care, makes sure that the cryptography is solid. So it's been an evolution for us to kind of step up to that gear.

A lot of competitors are looking at the same, but we definitely wanted to prioritize that, make sure the data is stored at rest in the best way possible and try to make it almost uncrackable. Right, because that's a huge component of identity security and really a lot of different security if we think about it. Now, a lot of organizations are adopting more cloud services, APIs, automation, AI workflows like you mentioned earlier which is really making identity even more critical as we seeing Where do you see identity security evolving over the next few years And what should organizations be doing now to stay ahead of that Yeah, so posture control is really important.

A lot of teams scale up. They've got cloud services, SaaS products. They may purchase and procure things over the next few years. Each one of those layers require authentication and control and management.

So the way we're trying to position ourselves is looking at how do we plug into those ecosystems so that we can also offer the same kind of management techniques. Can I rotate a password after you use it? Can I rotate a password on a schedule every time that we call on that? So each week, the key is being generated again.

That way, the user doesn't really have a foothold in any way. There's not a lot of standing privileges, even as we look at those SaaS products. So cloud accounts are a big one. That's where most servers might be running, most databases may be running.

And when I connect to a service using one of those keys, what does that key mean? Does that mean that I have admin control? And can I just copy and paste those keys locally and kind of bypass the system? Or once I use it, is it useless again?

and managing that at an automated way allows them to kind of scale that up. Some companies have hundreds of AWS accounts, so the problem really gets out of hand quickly. So having a system that you can say, rotate this password after I use it, rotate this key after I use it is incredibly helpful for that problem. Right.

Well, that's huge. And Jeremy, thank you so much for joining me today and sharing your insights. Before we close out, is there anything else you'd like to share with our audience? Yeah, really excited to see the progress that Keeper's made over the year.

We are actively looking at the next layer of threat detection and security. That's an area that I get to focus on heavily. So one of the cool things about managing these privilege sessions and identities is that we can stick on a little bit of user behavior analytics and explore using generative AI, as well as some deep learning methods to identify patterns and summarize that activity better. The hardest part is when an admin has a thousand people with 10 to 20 identities each.

That's a really tough problem to manage. Some companies only have a small team. So we try to do as much as we can to show keep or save the day, identify those points that maybe bad actors are working, maybe intentional behavior is happening, but it's not within the window of operations that you expect. So we're building all sorts of cool tooling around the use of AI to generalize some of that information, try to make the alerts and the reporting process much easier.

So I'm looking forward to seeing more of that stuff coming out from Keeper shortly. I love that. Well, thank you for that sneak preview of what's to come. For anyone listening, if you want to learn more about the incredible work Keeper Security is doing, go ahead and check out their website at KeeperSecurity.

com. That is K-E-E-P-E-R Security.com. Jeremy, thank you so much for joining me today.

You've been listening to Cyber Defense Radio. Stay tuned next time for another amazing and informative episode. CyberDefenseRadio.com is proudly part of the Cyber Defense Media Group, where InfoSec knowledge is power.

Cyber Defense TV and Cyber Defense Radio have launched 24x7x365 live streams. Visit them online today at Cyber Defense TV and Cyber Defense Radio with your host and cybersecurity champion and my good friend. Annabelle Thomas.