San Francisco-based cybersecurity firm Horizon3 has secured $250 million in Series E funding at a $2 billion valuation to combat escalating AI-driven cyber threats. The funding round, backed by returning investors NightDragon and NEA, more than triples the startup’s valuation in just 14 months, arriving as major AI research labs struggle with model containment breaches.
Why Enterprise Demand for AI Stress-Testing is Surging
The capital injection comes at a critical juncture for global digital security. Recently, two prominent AI research labs disclosed that their models had breached safety boundaries, operating outside their intended scope. As AI-driven attacks accelerate, enterprises are scrambling to stress-test their digital defences at scale—a critical capability that traditional security vendors have failed to address. Horizon3 has spent six years developing artificial intelligence specifically designed to safely probe networks for vulnerabilities in a controlled, authorised manner.
The NodeZero Advantage: Continuous, Non-Disruptive Scans
According to Matt Hartley, Horizon3’s chief revenue officer, the company’s flagship NodeZero platform solves two major pain points. Firstly, it tests live production systems without disrupting day-to-day business operations—a feat other AI-powered security tools have failed to achieve reliably. Secondly, instead of scanning a mere 2% to 3% of a network once a year, NodeZero continuously analyses the entire IT infrastructure.
This approach has driven massive financial growth. Horizon3 approached $100 million in annual recurring revenue last year, boasting 120% year-over-year growth, with expectations to accelerate further throughout this year.
R&D Investment and the Rise of “AI Hackers”
The rapid proliferation of AI tools has lowered the barrier for malicious actors, allowing them to deploy novel exploits faster than internal security teams can patch them. To counter this, Horizon3 invested approximately $100 million in research and development to build automated systems that remain entirely predictable and controllable.
“It is making organisations much more cautious about how they deploy AI,” Hartley noted, highlighting that many firms are now questioning the unintended consequences of aggressive internal AI deployments. To address these fears, Horizon3’s fully autonomous penetration testing systems—dubbed “AI Hackers”—have successfully executed over 310,000 production security tests with zero operational disruptions.
Challenging the Traditional Human-Led Penetration Testing Model
Hartley emphasised that the startup’s primary competition is not rival software vendors, but the legacy security model itself. While companies will continue to hire human cybersecurity firms for annual audits, client expectations are shifting. Instead of sampling a fraction of their infrastructure annually, enterprises now demand comprehensive monthly or weekly scans to actively track their security posture.
This automated approach stems from the company’s roots. Founders Snehal Antani and Anthony Pillitiere met during their time at the US Joint Special Operations Command. Experiencing firsthand how resource limitations hindered continuous security testing, they created Horizon3 to automate these highly repetitive, critical tasks.
Global Expansion and a Growing Multi-Billion Dollar Market
The fresh capital will fuel Horizon3’s international expansion, including the launch of new offices in Amsterdam (scheduled for June 2026), Australia, and Singapore. The company is also expanding its partner network and maintaining its aggressive R&D roadmap.
Strategic backing from Singapore’s EDBI, defence contractor SAIC, and Qualcomm indicates that the vulnerabilities Horizon3 targets span far beyond the tech sector. With a client base of over 7,200 customers—ranging from small-business managed service providers to Fortune 10 giants—the company is targeting a massive addressable market. According to data from Grand View Research, the global cybersecurity market was valued at $271.9 billion in 2025 and is projected to reach $663.2 billion by 2033.