Corma Secures $60 Million to Build Defensive AI Lab
San Francisco, Monday, 10 August 2026.
Corma raised $60 million in seed funding to launch a defensive AI lab, aiming to bridge the critical performance gap where AI attackers currently outperform security defenses.
Addressing the Asymmetry in Cybersecurity
The fundamental challenge in modern digital defense lies in a stark performance imbalance, where AI-driven offensive capabilities are rapidly outstripping human-led defenses [1]. In simulations conducted by Corma across realistic corporate environments, AI models acting as attackers achieved a success rate of 88%, whereas the same models acting as defenders managed to detect threats only 12% of the time [1]. This performance gap, represented by a difference of 76 percentage points, underscores the vulnerability of traditional security paradigms [1]. General-purpose AI models, while proficient at code reasoning, are often not optimized to parse the complex security data analysis—such as audit logs and network flows—required for robust defense [1].
A Specialized Workforce for Digital Defense
To counter these sophisticated threats, Corma is building the industry’s first foundation model designed specifically for defensive cybersecurity [1]. Founded in 2025 and operating out of San Francisco and Tel Aviv, the company has assembled an interdisciplinary team that unites frontier AI researchers from Google and DeepMind with elite cybersecurity specialists from Israel’s Unit 8200 [1]. Rather than relying on general-purpose models, Corma’s platform aims to deliver a complete, AI-powered defensive workforce capable of matching the speed and sophistication of modern automated attackers [1]. This tailored approach aims to level the playing field, providing enterprise security teams with the autonomous tools needed to defend critical systems at scale [1].
Rapid Deployment and Real-World Impact
Since its launch approximately six weeks prior to the funding announcement, around June 28, 2026, Corma’s AI technology has been rapidly adopted by Fortune 100 and Fortune 500 organizations [1]. These deployments span critical sectors, including healthcare, financial services, energy, critical infrastructure, and retail [1]. Early operational data reveals that the integration of Corma’s AI workforce has led to a 94% reduction in threat response times alongside a 15-fold increase in security coverage [1]. By automating the detection and mitigation of complex cyber threats, the system helps organizations secure their operations without the delays associated with manual human intervention [1].
Strategic Backing and Geopolitical Implications
The $60 million seed funding round, announced on August 10, 2026, was led by Sequoia Capital, with participation from Khosla Ventures and Coatue [1]. Financial backers and industry experts emphasize that the implications of autonomous cyber threats extend far beyond corporate balance sheets [1]. Vinod Khosla, founder of Khosla Ventures, noted that as cyberattacks become increasingly autonomous and scalable, the stakes move beyond data and finance to impact national security, geopolitical stability, and essential services like healthcare and critical infrastructure [1]. With this fresh capital, Corma is positioned to accelerate its research and deploy defensive AI systems capable of protecting vital global networks from highly sophisticated automated adversaries [1].