#United States #Mountain View #AI Security #Cyberhaven #Shadow AI

Cyberhaven Tackles the Rapid Growth of Shadow AI

Cyberhaven, a leader in AI-native data security, has recently made headlines with its expansion of the Unified AI & Data Security Platform. This initiative aims to address the challenges posed by the significant rise in the use of autonomous agents within organizations. Recent research indicates a staggering 509% jump in enterprise adoption of endpoint-based AI-native applications over the past year. This surge presents an urgent governance challenge as shadow AI systems evolve, functioning beyond the visibility and control of enterprises.

Understanding the Surge of Shadow AI

At the forefront of this transformation is an array of AI agents that operate autonomously and can access sensitive data. These agents, designed to improve productivity, also introduce unprecedented risks due to their lack of oversight within existing governance frameworks. According to Nishant Doshi, CEO of Cyberhaven, AI has transitioned from merely generating content to actively executing tasks that involve significant access to enterprise data and systems. He expresses the need for security measures that can operate in real time at the moment these AI agents take action, not merely on what users input into AI prompts.

Cyberhaven's Innovative Solutions

To combat the challenges posed by shadow AI, Cyberhaven has introduced several key features within its Agentic AI Security framework:

1. Discovery: This feature provides a continuously updated inventory of AI agents, GenAI applications, and Multi-Cloud Platform (MCP) servers. It evaluates shadow agents running locally on endpoints, providing Risk IQ scores across five vital dimensions to help organizations assess potential threats.

2. Observability: Cyberhaven reconstructs the entire lifecycle of agent interactions, tracking data access, invoked tools, actions taken, and conversational context. This holistic view allows for the automatic surfacing of violations that span multiple steps in the interaction process.

3. Controls: The new controls feature introduces runtime policy guardrails that guide users toward compliance by providing clear explanations rather than simply blocking unauthorized actions. This approach promotes better understanding among users about expected behaviors.

4. Data Lineage: A critical aspect of Cyberhaven's security measures is its data lineage capability, which connects each agent's actions to the data it interacts with. This connection provides context necessary for effective investigations, distinguishing mere alerts from actionable insights.

Saro Subbiah, SVP of Engineering at Cyberhaven, emphasizes the importance of visibility into the AI agents operating within corporate environments, asserting that this is essential for effective governance. With the new framework, Cyberhaven is setting a standard in recognizing and addressing the risks posed by AI agents.

Introducing the Analyst Plugin and Browser Extension

In addition to the foundational features of the Agentic AI Security solution, Cyberhaven has released two innovative tools:

• - Analyst Plugin: This tool transforms analysts into power users by embedding Cyberhaven's security intelligence directly into platforms like Claude Code and Codex. Equipped with over 40 pre-built security skills, the plugin streamlines incident triage, exposes vulnerabilities in GenAI applications, and enhances user risk profiling—all without interrupting analysts’ existing workflows.

• - Standalone Browser Extension: Extending data loss prevention capabilities to devices operating on ChromeOS and unmanaged endpoints, this extension monitors activities including uploads, downloads, and copy-paste actions. It implements content inspection and is managed via the same interface utilized for endpoint security.

As enterprises look to embrace AI solutions while ensuring robust security measures, Cyberhaven's expansions stand as a model for governance in the AI era. These tools not only protect sensitive data but also empower security teams to take proactive actions against the challenges of an increasingly AI-driven workplace.

For organizations wanting to learn more about these new offerings, Cyberhaven is hosting a live launch event—details can be found on their official website. Cyberhaven continues to safeguard sensitive data, reduce insider risks, and foster secure AI adoption, all while maintaining operational fluidity in business processes.