Onyx Security has raised $113 million in a Series B funding round to expand its enterprise platform for monitoring and controlling artificial intelligence agents. Bessemer Venture Partners led the financing, with participation from Cyberstarts, TCV, Conviction, FirstMark, Vintage Investment Partners, QuantumLight, and G Squared. The round brings the New York- and Tel Aviv-based company’s total funding to $153 million two years after its founding.
Rapid Growth After Emerging From Stealth
The financing follows a period of rapid commercial expansion for Onyx, which reported that its revenue has quadrupled since the company emerged from stealth four months ago. Onyx also said its technology now secures more than 1.1 million AI agents across enterprise customers and inspects more than 66 million AI sessions in real time. Its customer base includes Fortune 500 companies operating in banking, technology, insurance, energy, healthcare, and other industries where automated decisions can carry significant operational or regulatory consequences.
Building an Enterprise AI Control Layer
Onyx has developed what it calls a Secure AI Control Plane, designed to give security, governance, and infrastructure teams visibility into how AI agents operate across corporate environments. The platform discovers agents working through browsers, endpoints, software-as-a-service applications, and cloud systems, then evaluates their actions before those actions take effect. Using proprietary models, the company says it can monitor agent reasoning, identify unintended or malicious behavior, enforce organizational policies, and intervene without preventing businesses from adopting AI tools.
Addressing Risks From Autonomous Agents
The company is positioning its technology around the growing use of AI agents that can access sensitive systems and execute increasingly complex tasks with limited human involvement. As enterprises delegate activities such as software development, workflow automation, data analysis, and infrastructure management, security teams face the challenge of determining whether an agent’s actions are authorized, compliant, and safe. Onyx argues that conventional security products were not designed to govern autonomous actors capable of making decisions and operating across several digital environments at production speed.
Funding Product Development and Expansion
Onyx plans to use the new capital to train the next generation of its proprietary models and expand its enterprise go-to-market operations in the United States and other markets. The company will focus particularly on critical sectors such as financial services, energy, and healthcare, where errors by autonomous systems could contribute to outages, financial disruption, compliance failures, or risks to patients. The funding will also support Onyx as it deepens integrations with AI providers, following an integration with Anthropic announced in June and connections with dozens of other AI platforms.
Investor Support and Market Position
Bessemer Venture Partners said enterprises need stronger trust, accountability, and control mechanisms before AI agents can be deployed at scale across important business functions. Cyberstarts similarly identified agent security and governance as a potentially significant cybersecurity category as autonomous systems become embedded in critical workflows. Onyx co-founder and CEO Maxim Bar Kogan, who launched the company with co-founder Gil Elbaz, said the central objective is to ensure humans retain oversight as AI agents perform a larger share of enterprise actions.
The $113 million Series B gives Onyx substantial resources to compete in the emerging market for enterprise AI governance and runtime security. Its growth since leaving stealth indicates rising corporate demand for tools that can identify agents, examine their behavior, and stop harmful actions before they affect business systems. As organizations expand their use of autonomous AI, Onyx is betting that a dedicated control layer will become essential infrastructure for balancing faster adoption with security, compliance, and human accountability.