There’s a saying: the best defense is a good offense. That couldn’t ring more true than cybersecurity in 2026, where “The best defense is a good red team.”
This summer, OpenAI disclosed that its models had broken out of a cyber evaluation and breached Hugging Face’s servers using stolen credentials and zero-day vulnerabilities. Within days, Anthropic published that Claude models had reached the live systems of three organizations during testing, and Meta soon said one of its models had done the same. None of these were adversaries. They were models doing what they were asked a little too well, in environments that assumed a boundary would hold. If an evaluation can do that by accident, an attacker with the same capability can do it on purpose, at machine speed, against every surface an enterprise has.
That was all anyone at Black Hat could talk about this year, and it has changed the question security leaders are being asked. Boards no longer care whether the company ran its annual pentest. They want to know whether it’s secure right now, across applications, cloud, identity, and internal networks, and what to fix first if it isn’t.
And CISOs aren’t just thinking about the answer. They’re acting on it now, running pilots, moving budget, and buying, and they want the best rather than a checkbox. In the words of one CISO, what’s coming is security’s version of Game of Thrones’ Red Wedding, and there’ll be a line between the organizations that can find and fix fast enough and those that can’t. And nobody wants to be on the wrong side of it.
That’s why we’re excited to lead Armadin’s Series B.
We believe there is no better team than Kevin Mandia, Travis Lanham, Evan Peña, and David Slater to take this on. Kevin is one of the most notable figures in security: he founded Mandiant, the firm that defined modern incident response, and has been at the center of every major shift in how attackers operate since. Travis, Evan, and David bring deep experience building Google’s market leading security products and leading Mandiant’s red team, so they know offense and defense from the inside. This time, they’re building the autonomous security platform defenders need for the AI era.
If AI can find a way in at machine speed, the only way to know you’re safe is to pressure-test your own environment the same way, on your terms, before a real attacker gets the chance. That’s hard to do with a model built around human testers, where red team engagements and pentests are scoped in advance, staffed by scarce specialists, and run a few times a year against a fraction of the estate. The attack surface changes every time a team ships, and attacker tooling improves with every model release, so last quarter’s test is already describing an environment that no longer exists. The window between a vulnerability appearing and someone exploiting it is shrinking fast, and AI is shrinking it further. Testing has to run continuously, at the same scale attackers now operate.
As with many product shifts in the history of cyber, yesterday’s product, business model, and needs look almost nothing like today’s. That’s where Armadin comes in.
Armadin’s AI autonomous security platform turns offense into something defenders control, from the first Hyperattack. An agentic attacker swarm, built on decades of frontline adversary tradecraft, reasons, plans, and adapts the way a real adversary would across external assets, cloud, identity, internal networks, and applications. What comes back is decision-grade proof: validated kill chains that show reachability, exploitability, and blast radius, followed by proactive remediation to close them. Because the platform maintains a live graph of the attack surface and gets sharper with every attack and environment, security teams spend their time fixing real problems instead of triaging noise. That kind of precision comes from a company built only for security.
Armadin Red is just the beginning. Kevin and the team have spent their careers staying a step ahead of attackers, and now they’re building the platform to help every security team do the same.
- State of Markets II David George
- OpenAI Understands Something Important and Rare David George
- Expanding the a16z Growth Fund and Platform David George and Raghu Raghuram
- The Machine Age Fund Ben Horowitz, Martin Casado, Raghu Raghuram, David Ulevitch, and David George
- Is AI a Bubble? | Gavin Baker on Data Centers, GPUs, and the AI Economy Gavin Baker and David George
- Investing in Gimlet Raghu Raghuram, Sarah Wang, Shangda Xu, and Stephenie Zhang
- Investing in Exa Sarah Wang, Jennifer Li, Stephenie Zhang, and Jason Cui
- From “System of Record” to “System of Intelligence” Gio Ahern, Stephenie Zhang, and Alex Immerman
- Investing in Temporal Sarah Wang, Raghu Raghuram, and Stephenie Zhang
- Big Ideas 2026: The Agentic Interface Stephenie Zhang, Sarah Wang, and Marc Andrusko