Swarm of agents reshapes penetration testing as Armadine raises $255 million to thwart self-directed attacks

Traditional penetration tests that rely on hiring human teams to inspect systems periodically are no longer sufficient to keep pace with the new era, as the capabilities of autonomous models to exploit software vulnerabilities are accelerating. In this context, the companyArmadinespecializing in cybersecurity, founded by Kevin Manda, announced raising $255.5 million in a Series B funding round, bringing its valuation to over $2.5 billion. This round comes just six months after the company raised $190 million in a Series A last March, taking the total amount raised to $445 million in a short period, a clear indication of the scale of investment bets on self-defense solutions.

The financing round was led by Andreessen Horowitz and Accel, with broad participation from prominent investment funds including Bain Capital Ventures, Redpoint, ETV, Balistic Ventures, Google Ventures, In-Q-Tel, Kleiner Perkins, and Menlo Ventures. This investment momentum is based on the professional track record of founder Kevin Manda, known for founding Mandiant, which Google acquired in 2022 for $5.4 billion. However, the fundamental shift in his new venture is not linked solely to prior experience, but to the operational model he proposes for dealing with threat environments emerging from autonomous agent technologies.

Armadine's operational concept rests on reinventing corporate defense testing entirely to meet the requirements of the AI era, by providingnonstop agent swarmsto simulate attacks. Unlike the traditional approach that relies on seasonal penetration attempts carried out by external teams and delivers reports after days or weeks, the platform launches swarms of software agents that link disparate vulnerabilities into continuous attack chains to infiltrate systems. The primary goal is to help institutions and companies discover those gaps and remediate them instantly before hostile actors or even stray models that escape AI research labs exploit them using agent technologies to launch complex attacks.

This shift carries direct practical implications for information security leaders and digital infrastructure protection teams in the Gulf, Egypt, and the Arab Levant. For decades, critical sectors such as banks, telecom companies, and government agencies have relied on periodic penetration tests as a core standard for regulatory compliance and annual verification. However, the entry of offensive agent swarms imposesa change in the decisions and skills requiredwithin regional cyber-defense teams; static reports are no longer sufficient to protect infrastructure that is continuously targeted automatically. Technology leaders will need to redirect protection budgets toward continuous offensive auditing tools, and shift cybersecurity specialists' tasks from labor-intensive manual scanning to training and monitoring defensive agent swarms capable of anticipating complex intrusion chains at the same speed.

Armadine's rapid valuation jump to $2.5 billion in half a year confirms that cyber defense has entered a fully autonomous operating stage. If you manage a technology system in the region, the message is clear: future attacks will not wait for the next periodic audit schedule, and organizations that do not build their capability to test their systems with connected automated swarms will find themselves several steps behind attacks launched at computing-processing speed rather than human decision speed.