Zscaler launches AI security suite to tackle enterprise risk and visibility gaps
New capabilities aim to secure generative and agentic AI use across infrastructure, applications and data
Cybersecurity provider Zscaler has launched a suite of artificial intelligence security features designed to help enterprises protect and govern the expanding use of generative and agentic AI systems.
The company said most organisations lack comprehensive visibility into AI applications, models, agents and infrastructure, limiting their ability to assess data exposure and security posture.
In its newly published ThreatLabz 2026 AI Security Report, Zscaler found critical flaws in 100% of enterprise AI systems analysed, with most vulnerable to compromise within 16 minutes.
The new platform provides an inventory and dependency map spanning generative AI services, embedded AI in software-as-a-service applications, development environments, model control plane servers, agents, models and supporting infrastructure.
“AI is changing how businesses operate, but traditional security approaches were not designed to secure AI,” said Jay Chaudhry, chief executive, chair and founder of Zscaler.
“Business leaders are looking for a comprehensive solution, not more point products. At Zscaler, we’re providing the security necessary for leaders to move forward with confidence and embrace the full spectrum of AI.”
The suite supports three enterprise use cases: AI asset management to discover unauthorised tools and map data access; secure access to AI using Zero Trust controls, inline inspection and prompt classification; and protection of AI infrastructure and applications through automated red teaming, prompt hardening, runtime guardrails and continuous posture monitoring.
“The industry is currently struggling with a massive visibility gap because AI traffic doesn’t behave like traditional web traffic,” said Zeus Kerravala, principal analyst at ZK Research.
Related reading
- OpenAI lifts lid on Codex engine with technical deep dive
- OpenAI scales PostgreSQL to millions of queries per second for ChatGPT users
- AI spending momentum at Davos points to 2026 inflexion year
Zscaler said the platform aligns with governance frameworks such as the United States National Institute of Standards and Technology AI Risk Management Framework and the European Union AI Act.
The system integrates with providers including OpenAI, Anthropic, Amazon Web Services, Microsoft and Google, and introduces new defences including a model control plane gateway for secure automation and AI deception tools.
The Recap
- Zscaler unveiled an AI security suite to protect enterprise AI.
- ThreatLabz found critical flaws in 100% of systems analyzed.
- Zscaler is expanding defense capabilities with a new MCP gateway.
