New features within the Zscaler platform enable organizations to gain insight into AI applications and securely build, deploy, and use them.
Zscaler is expanding its AI security offering with new capabilities that help organizations securely deploy and manage AI applications. The focus is on better insight, control, and governance when using generative and agent-based AI.
read also
What is Zscaler and how does one pronounce it?
Security for the entire AI landscape
The new features within the Zscaler AI Security Suite provide organizations with an overview of all AI elements within their IT environment. This includes GenAI tools, embedded AI in SaaS services, development environments, models, agents, and underlying infrastructure. By mapping these components, companies can better assess risks and control data access.
According to Zscaler, traditional security solutions do not provide sufficient protection for AI use. AI traffic uses different protocols and exhibits non-human behavior, which complicates detection. The new suite combines asset discovery, access control, behavioral research, and risk assessment. This allows companies to roll out AI faster without compromising security or policy.
Three security pillars and support for regulations
The suite focuses on three scenarios: insight into AI assets, secure access to AI, and protection of AI infrastructure and applications. CISOs and IT teams get tools to detect ‘hidden’ AI and analyze data usage. At the same time, zero trust controls and automated inspection enable safe use. For development teams, there are security measures for the entire AI lifecycle.
Zscaler also supports compliance with frameworks such as the NIST AI Risk Management Framework and European AI legislation. There are also collaborations with OpenAI, Anthropic, AWS, Microsoft, and Google. Finally, a new gateway with AI Deception is coming to detect and counter attacks on models.