AI security firm Irregular announced on Wednesday that it has raised $80 million in a funding round led by Sequoia Capital and Redpoint Ventures, with participation from Wiz CEO Assaf Rappaport. The round values the startup at $450 million, according to a source familiar with the deal.
Irregular, formerly known as Pattern Labs, has become a key player in AI model evaluations. Its work is cited in security testing for Claude 3.7 Sonnet, OpenAI’s o3, and o4-mini models. The company also developed SOLVE, a widely used framework that scores a model’s ability to detect vulnerabilities.
But the startup’s ambitions are growing. Instead of only analyzing existing risks, Irregular is building systems to catch emergent risks and behaviors before they appear in real-world use.
“Our view is that soon, a lot of economic activity is going to come from human-on-AI interaction and AI-on-AI interaction,” said co-founder Dan Lahav. “And that’s going to break the security stack along multiple points.”
To prepare for that future, Irregular has designed network simulations where AI models act as both attackers and defenders. This setup allows researchers to see how new AI systems hold up under pressure before deployment.
Security has become a hot-button issue in the AI industry, with leading labs overhauling their own practices amid growing risks of corporate espionage and misuse. At the same time, the ability of AI to find and exploit software vulnerabilities has created new challenges for defenders.
Related: Meet Keplar: The AI Voice Startup Changing Customer Research Forever
“If the goal of the frontier lab is to create increasingly more sophisticated and capable models, our goal is to secure these models,” Lahav said. “But it’s a moving target, so inherently there’s much, much, much more work to do in the future.”
