Zenity Adds Agentic Browser Protection, LLM Defense Tools

This week, Zenity announced a major expansion to its AI security stack, introducing new capabilities designed to help security teams understand, investigate, and govern the rapidly growing universe of AI agents, assistants, and autonomous browsers.

Release delivers deeper visibility, agentic browser coverage and open source defense tools

For resellers and integrators supporting enterprise customers scaling AI adoption, the update targets one of the sector’s biggest pain points: a lack of visibility into how AI-driven activity unfolds across identities, workflows, and environments.

“With this release we are giving security teams something they have never had before, real visibility into intent,” said Ben Kliger, co-founder and CEO of Zenity, in a statement. “Our new Correlation Agent does not just detect signals, it interprets them. It understands what an agent is trying to do, by connecting every signal, data point and insight that the Zenity platform collects and generates throughout the agent lifecycle into a single coherent story.”

“This is a game changer for AI security, especially as organizations embark on their goals towards 1 billion agents,” Kliger added. “By transforming scattered signals into high confidence security narratives, we are eliminating guesswork, accelerating investigations and giving teams the clarity they need to operate safely at massive scale.”

Intelligence layer correlates signals into actionable narratives

Traditional security tools generate alerts but rarely provide the story behind them. Zenity’s new Issues capability acts as an intelligence layer that correlates posture findings, runtime anomalies, identity relationships, and graph-based insights into unified, high-confidence incidents. 

Instead of sorting through fragmented signals, teams receive a coherent narrative that explains what happened, why it happened, and what systems or data were affected.

At the center of this release is the Correlation Agent, which interprets AI behavior rather than simply flagging events. Zenity says the tool surfaces manipulation attempts, clarifies an agent’s intent, and dramatically reduces guesswork—an increasingly urgent need as organizations approach the “1 billion agents” operating within enterprises that Zenity predicts will be in place in the near future. 

For partners, this means faster investigations, clearer root-cause analysis, and the ability to provide customers with more immediate, contextualized responses to emerging threats.

Expanded coverage for agentic browsers include security for ChatGPT Atlas and others

Zenity has also expanded protection for agentic browsers, including ChatGPT Atlas, Perplexity Comet, and Dia. These tools automatically read content, take actions within authenticated sessions, and blur the line between human and agent activity. 

Through Zenity’s device agent, organizations can now discover agentic browsers in use, monitor autonomous actions, apply DLP controls, and detect intention-driven anomalies in real time. For MSPs tasked with securing shadow AI adoption, this unified governance provides the oversight and policy consistency needed across the entire fleet of enterprise AI tools.

Safe Harbor open source project strengthens LLM defense

Zenity Labs is also introducing Safe Harbor, a new open-source tool designed to counter emerging LLM manipulation techniques, such as data structure injection and structured self-modeling attacks. 

Safe Harbor gives agents a “safe action” they can automatically invoke when they detect harmful workflows or suspicious inputs, allowing AI systems to pivot away from potentially malicious instructions. Developers can use the project during the build process to reduce runtime exploitation risks before deployment.