Zenity Raises $125 Million Series C to Strengthen AI Agent Security and Governance

AI security and governance platform Zenity has announced a $125 million Series C funding round led by Norwest. The investment signals escalating enterprise concern over the security risks associated with autonomous AI agents that access sensitive data, invoke tools, execute workflows, and drive business decisions.

Joining lead investor Norwest, new strategic backers Qumra Capital, SoftBank Vision Fund 2, Hitachi Ventures, and LG Technology Ventures participated in the round. Existing investors Vertex Ventures, Third Point Ventures, DTCP, and Intel Capital also re-invested.

Zenity Raises $125 Million for AI Agent Security

Unlike traditional AI security tools that focus strictly on model protection, static guardrails, or input prompt filtering, Zenity targets the runtime actions of autonomous AI agents operating inside enterprise environments.

AI agents are rapidly becoming embedded across software-as-a-service (SaaS) platforms, cloud infrastructure, developer workflows, web browsers, and core business tools.

These agents connect directly to critical enterprise resources, including internal email, cloud storage buckets, customer relationship management (CRM) records, calendars, source code repositories, and productivity suites.

While this automation offers massive efficiency gains, it creates uncharted attack surfaces. A manipulated or hijacked agent can inadvertently exfiltrate confidential data, trigger unauthorized financial or API workflows, or misuse connected tools.

Implementing robust AI security frameworks allows organizations to establish deterministic governance over these non-deterministic agentic systems.

Zenity’s platform provides real-time visibility, preventive guardrails, and behavioral governance for agentic environments.

The technology analyzes an agent’s intended actions in real time, deterministically allowing, modifying, or blocking requests before execution. This approach transitions AI defense from post-incident log analysis to active threat prevention.

As detailed in the funding disclosure on Zenity’s official newsroom, the platform supports both commercial off-the-shelf agents and custom agentic frameworks.

Aligning agent guardrails with a GenAI security baseline helps organizations mitigate prompt injection and unauthorized agency across multi-vendor AI deployments.

Deployment Category	Supported Frameworks & Ecosystems
Enterprise SaaS Agents	Microsoft Copilot, ChatGPT Enterprise, Google Gemini
Developer Assistants	Claude Code, Codex, Cursor
Custom Agent Frameworks	AWS Bedrock & AgentCore, Microsoft Foundry, Google Vertex AI

Zenity is trusted by major enterprises across financial services, healthcare, pharmaceuticals, manufacturing, energy, and technology, including Fortune 500 and Global 2000 leaders such as SoftBank Corp.

The company has tripled its annual recurring revenue in each of the past two years and expects to maintain this growth trajectory.

Zenity Labs, the company’s threat research arm, actively contributes to industry standards and frameworks like OWASP and MITRE ATLAS:

• AgentFlayer Research: Highlighted zero-click exploit scenarios where threat actors hijack enterprise AI agents, manipulate multi-step workflows, and exfiltrate sensitive records without direct user interaction.
• Indirect Prompt Injection: Demonstrated real-world attack vectors involving malicious calendar invites, poisoned documents, and vulnerable tool integrations.

Deploying specialized autonomous AI security platforms enables security teams to proactively test and harden these complex agentic execution paths.

With a global workforce exceeding 230 employees, Zenity maintains R&D operations in Tel Aviv alongside go-to-market and executive management teams in New York.

The $125 million Series C capital injection will accelerate expansion across Europe and Asia-Pacific, scale Zenity Labs research operations, and expand preventive security controls for enterprise AI deployments.

[Live Webinar] Join Elastic & UnderDefense to learn how small security teams can unify AI visibility and agentic response into one operating model -> Register Now