Corma Raises $125M Series A to Fight AI-Powered Cyber Threats
AI cybersecurity startup Corma hits $1B valuation just three months after launch, backed by NEA and others, as security threats accelerate.
AI cybersecurity startup Corma hits $1B valuation just three months after launch, backed by NEA and others, as security threats accelerate.
Corma, an AI-focused cybersecurity startup, has just closed a $125 million Series A funding round at a $1 billion valuation, led by venture firm NEA. The company’s rapid ascent speaks volumes about the urgency of the moment: we’re living in an era where artificial intelligence doesn’t just create new products, it creates new attack surfaces.
The timing couldn’t be more telling. Just three months after Corma emerged from stealth in August, the company found itself riding a wave of panic and opportunity. On October 21, a perfect storm of security incidents hit the industry simultaneously. A critical flaw was discovered in PHP, one of the world’s most widely-used web programming languages. The same day brought news of another vulnerability in Node.js, the popular open source JavaScript runtime. LiteLLM, a proxy tool for routing requests to multiple AI language models, also disclosed a security incident. All of this happened at once.
That single day illustrated exactly why tech companies are suddenly so eager to fund cybersecurity solutions. When multiple infrastructure layers fail in coordinated fashion, the attack surface becomes almost impossible to defend manually.
Corma certainly isn’t alone in recognizing this opportunity. Kevin Mandia, the legendary cybersecurity veteran and former head of US Air Force Cyber Command, launched Armadin with this exact mission. Established tech giants like Anthropic, OpenAI, and Microsoft are all investing heavily in AI-powered security. Dozens of startups have emerged to address different angles of the problem.
Yet Corma’s rapid funding suggests investors believe there’s meaningful differentiation to be found. The company positions itself specifically as a frontier AI research lab dedicated to protection. That positioning matters because cybersecurity, unlike many software categories, can’t be solved through incremental product improvements alone. The threat landscape evolves faster than traditional defense mechanisms can adapt. Frontier AI research gives Corma the theoretical flexibility to stay ahead of emerging attack patterns.
What makes this moment particularly intriguing is the paradox at its core: AI has both created the problem and enabled the solution. The same AI capabilities that empower more sophisticated attacks also enable more sophisticated defenses. The question becomes less about whether AI can help defend infrastructure and more about whether the offense-defense cycle accelerates faster than anyone can manage.
Corma’s total funding now stands at $185 million, with backing from marquee names like Sequoia Capital, Khosla Ventures, and Coatue. That’s substantial validation from investors who typically don’t move this fast unless they see genuine market urgency.
The October 21 incident cascade provides the perfect case study in why this market exists. When multiple critical infrastructure components fail within hours, CISOs face an impossible choice: patch aggressively and risk breaking dependent systems, or move cautiously and accept elevated risk. AI-powered defense systems that can simulate threat scenarios and recommend interventions with higher confidence become invaluable.
But here’s the uncomfortable truth underlying all this excitement: no cybersecurity solution, AI-powered or otherwise, can make infrastructure truly secure. What good solutions can do is reduce the window of exposure and increase the cost of successful attacks. They buy time for teams to respond. They force attackers to expend more resources to achieve their objectives.
Corma and its competitors are building tools that might do exactly that. Whether they can do it sustainably while the threat landscape continues its relentless evolution remains the real test. The venture capital flowing into this space suggests investors believe the answer is yes. But the October incidents also suggest we haven’t seen the worst of what coordinated infrastructure failures might look like.
Source: TechCrunch