Chicago-based cybersecurity startup Empirical Security has raised $25 million in Series A funding led by Brightmind Partners, with participation from existing investors Costanoa Ventures and Hyde Park Angels. The round brings the company’s total funding to $37 million, following a $12 million seed round in July 2025.
What the Company Does
Empirical builds AI and machine learning models that help security teams predict which vulnerabilities are actually likely to be exploited, rather than relying on generic, one-size-fits-all risk scores. The company’s approach centers on two products: Foundation, a global model that monitors more than 18,000 exploited CVEs to identify emerging threat patterns, and Radiant, a custom predictive engine trained on each organization’s own data and environment to surface the threats most relevant to it.
The new capital will go toward accelerating development and deployment of both products.
Founders With a Track Record
Empirical is led by CEO Ed Bellis, CTO Michael Roytman, and Chief Data Scientist Jay Jacobs, who previously worked together at Kenna Security, a company they built around the concept of risk-based vulnerability management before its acquisition by Cisco in 2021. Jacobs and Roytman are also among the creators of the Exploit Prediction Scoring System (EPSS), an open-source framework for estimating the likelihood that a given vulnerability will be exploited.
Why It Matters
Generative AI is expanding both the scale and sophistication of cyberattacks, putting pressure on security teams to move from reactive vulnerability patching toward predictive prioritization. Empirical’s pitch is that static, generic risk scores are no longer sufficient, and that models trained on an organization’s specific infrastructure and data can better separate genuine risk from background noise.
Leave a Reply