.jpg)
As enterprises scale the deployment of AI agents and autonomous tools, the security challenges are evolving rapidly. Zenity, a security and governance platform for AI agents, has announced a significant expansion of its capabilities to address these new complexities. The update introduces an intelligence layer for correlating AI-driven security incidents, extends coverage to "agentic browsers," and releases an open-source tool designed to help developers evaluate and mitigate LLM manipulation risks.
Zenity has expanded its AI security platform with new incident intelligence and correlation capabilities.
The platform now offers security coverage for "agentic browsers" like ChatGPT Atlas, Perplexity Comet, and Dia.
A new open-source tool, "Safe Harbor," helps developers evaluate and mitigate LLM manipulation risks during the build phase.
The "Correlation Agent" interprets signals to build coherent security narratives, explaining an AI agent's intent and actions.
The update aims to reduce guesswork for security teams and provide visibility into previously opaque agentic browser activity.
Zenity positions these features as critical for organizations scaling toward widespread AI agent adoption.
A core challenge in AI security is the volume of disparate alerts that lack context. Zenity's new Incident Intelligence capability, powered by a "Correlation Agent," aims to solve this by linking posture findings, runtime anomalies, and identity data into coherent narratives. This gives security teams a story explaining what an agent attempted to do, why, and the impact, moving beyond simple signal detection to intent interpretation. CEO Ben Kliger stated this provides "real visibility into intent," accelerating investigations.
The platform now addresses the emerging risk of agentic browsers—tools that autonomously browse the web and interact with authenticated sessions on a user's behalf. Zenity highlights ChatGPT Atlas, Perplexity Comet, and Dia as examples. These tools create a "shadow AI" problem where security teams cannot distinguish between human and agent activity. Zenity's device agent discovers these browsers, monitors their autonomous actions, applies data loss prevention policies, and detects intention-based anomalies to close this visibility gap.
Through its research arm, Zenity Labs, the company is releasing Safe Harbor, an open-source tool based on research into data structure injection and self-modeling attacks against LLMs. It provides a "safe action" an agent can call when it identifies a harmful instruction, allowing it to pivot away from a malicious workflow instead of completing it. This tool is designed for developers to integrate safety measures during the AI application build process, not just in runtime monitoring.
These launches reinforce Zenity's focus on securing AI based on its unique behaviors—autonomous decision chains, tool invocation, and interaction with enterprise data. By correlating incidents, governing new agentic application categories, and contributing to the broader security community, Zenity is building a comprehensive defense-in-depth approach tailored for an agent-driven future. Kliger emphasized the need for this as organizations work towards goals of deploying "1 billion agents."
For security teams, these advancements promise to reduce alert fatigue, provide critical context for AI incidents, and extend governance to a new class of AI-powered applications that operate outside traditional application perimeters. It represents a proactive step in defining the security model for the next wave of enterprise AI adoption.
About Zenity
Zenity is the first security and governance platform purpose-built for AI agents - spanning SaaS, home grown platforms (Cloud), and end-user devices (Endpoint). Trusted by Fortune 500 enterprises, Zenity helps security teams confidently adopt AI by delivering defense in depth with full-lifecycle coverage: from agent discovery and posture management to real-time detection, inline prevention, and response. With an agent-centric approach that prioritizes how agents behave, what they access, and which tools they invoke, Zenity eliminates blind spots and enforces consistent policy and controls across environments so organizations can innovate with AI, without compromising security.
TechIntelPro empowers B2B technology decision-makers with the latest industry trends, deep insights, leadership perspectives and actionable intelligence to drive both organizational and professional milestones—and thrive on the tech pulse.