Volume 343, Issue 3 · 6 articles
Citing the eMudhra CA compromise, Zscaler argues TLS inspection protects enterprise apps that cannot react to PKI incidents as fast as browsers.
Zscaler says the recent eMudhra certificate authority compromise exposes an overlooked challenge.
Modern browsers can quickly revoke trust in compromised CAs, but many enterprise applications and devices cannot respond to PKI incidents as fast.
The authors argue that certificate trust is not absolute and that the enterprise internet extends well beyond the browser.
They explain how TLS inspection in Zscaler Internet Access can act as a safety net by letting organizations change trust quickly.
Zscaler CSO Deepen Desai recaps August-September ThreatLabz research, including AI-assisted attacks, ransomware and AI readiness reports.
Zscaler Chief Security Officer Deepen Desai recaps August and September milestones for Zscaler and ThreatLabz.
Highlights include the annual ThreatLabz 2026 Ransomware Report, the 2026 Frontier AI Readiness Report and the Agentic SecOps launch.
The roundup covers AI-assisted attacks in Japan and South Korea that used AI for reconnaissance, vulnerability discovery and exploitation.
It also covers Operation RapidRust and other new malware and notable campaigns.
Zscaler helps enterprises see how Cursor is used on managed devices and apply security controls to AI coding agent activity.
Zscaler says AI-assisted development with Cursor is becoming part of how enterprises build software.
A Cursor agent's workflow can extend beyond the editor to access files, run commands, call MCP tools, connect to external services or install software.
Zscaler helps organizations understand how Cursor is used on managed devices and apply the right security controls.
The post explains how this extends AI governance to the endpoint to support secure enterprise adoption.
Zscaler argues national security agencies should operationalize Security Classification Guides as zero trust data controls using DSPM.
Zscaler says national security organizations share a common problem: security guidance, data and mission needs exist, but the control path between them is manual.
Security Classification Guides explain what makes information sensitive and when combinations of facts become more sensitive, yet they usually sit in static documents.
Meanwhile, mission data flows through file shares, databases, data lakes and SaaS platforms.
The authors explain how to operationalize these guides for zero trust data decisions and where Zscaler DSPM fits.
Zscaler explains how its AI Guard and Sentry extend NVIDIA OpenShell's agent boundaries to everything an AI agent can reach.
Zscaler EVP Dhawal Sharma says CISOs have moved from asking whether to allow AI to asking how to safely move agents into production.
The post describes Zscaler's support for the NVIDIA OpenShell reference architecture for safe AI agents.
NVIDIA draws the boundary around the agent, and Zscaler extends it to everything the agent can reach.
Sharma explains how Zscaler AI Guard and Sentry work with OpenShell and how in-cluster detection will become prevention.
Zscaler ThreatLabz finds AI-assisted attacks combining the ARTEX pentest framework with AI coding assistants targeted organizations worldwide.
Zscaler ThreatLabz investigated AI-assisted cyberattacks that combined the ARTEX penetration testing framework with AI coding assistants and custom automation scripts.
Initial reports focused on attacks against South Korean financial institutions.
ThreatLabz found the campaign went far beyond South Korea, targeting organizations around the globe across multiple industries.
The post examines the operator's working environment, infrastructure and attribution, with IOCs and a MITRE ATLAS mapping.