Onyx Security announced its launch alongside $40 million in funding as the company seeks to help enterprises safely deploy and manage autonomous AI agents across their organizations. The round was backed by Conviction and Cyberstarts, and the company positions its platform as a centralized security and governance layer for the rapidly emerging “agentic” era of artificial intelligence.
AI agents are increasingly being used by companies to automate tasks across departments ranging from engineering to customer service. However, these systems introduce new operational and security risks. Because AI agents can reason, access sensitive systems, and take actions independently, errors or hallucinations can lead to unintended outcomes. At the same time, adversaries can attempt to manipulate agents using malicious prompts, creating a new attack surface for enterprises.
Onyx Security is addressing this challenge by building what it describes as a secure AI control plane. The platform is designed to discover AI agents operating inside an organization, monitor the reasoning steps they take, and approve, modify, or block their actions in real time. By supervising how agents interact with enterprise systems, the company aims to allow organizations to adopt AI agents while maintaining governance, security, and compliance.
The company’s platform is intended to support multiple teams across an enterprise. Security teams gain visibility into how AI agents operate and receive tools for posture management and runtime protection. Governance teams can use the platform to enforce regulatory and internal policy requirements, while infrastructure teams can manage agent deployments and optimize for performance factors such as cost, accuracy, and latency.
Onyx uses its own supervisory agents and proprietary AI models to analyze how other agents reason and act. These models can intervene when necessary by blocking actions, requiring human approval, or redirecting an agent toward a safer path. The goal is to transform autonomous AI agents into controlled participants within enterprise workflows rather than unsupervised digital workers.
According to the company, several Fortune 500 organizations are already using the platform. Onyx has also grown to more than 70 employees across Israel, the United States, and Canada.
The newly raised capital will be used to expand product and engineering teams, develop additional AI models, and scale the company’s go-to-market operations as demand for AI governance and security tools increases.
The company was founded by Maxim Bar Kogan and Gil Elbaz. Bar Kogan previously served as a cybersecurity leader and technology executive and was a member of Unit 8200 of the Israel Defense Forces. Elbaz is an AI researcher who previously reported to the CTO of Nvidia and was part of one of the IDF’s AI research units.
As enterprises increasingly integrate autonomous AI systems into critical operations, Onyx aims to provide the infrastructure needed to ensure those systems operate safely, securely, and within defined organizational policies.
KEY QUOTES:
“Every enterprise is becoming an agent operator — whether they planned to or not,” “Agents are given access to the most critical systems in the enterprise, but what are our guarantees they will not make serious mistakes or get compromised? How do we decide what systems are off limits for agents, or what are the requirements to access them? The safe adoption of AI agents requires security from attacks, as well as ensuring agents don’t make critical mistakes.”
Maxim Bar Kogan, Co-Founder And CEO Of Onyx Security
“Maxim and Gil saw the need for an AI control plane before the market did. While most of the industry was still focused on data loss in chatbots, they predicted that the risk surface of agents was going to be critical. Onyx is the control plane all enterprises will need as they scale to thousands of agents.”
Sarah Guo, Founder Of Conviction Partners
“The shift of organizations toward a world in which AI systems become part of their operational infrastructure creates an entirely new attack surface. These are not just software tools to be protected – they are systems that make decisions, access sensitive information, and are integrated into critical business processes. From the very first moment, it was clear to us that Maxim and Gil understand the depth of this change and are building a solution designed for the AI era, rather than trying to adapt legacy tools to a new reality. We believe the Onyx team is exceptionally well-positioned to become a market leader in AI security for large enterprises.”
Hila Zigman, Partner At Cyberstarts
