What's New in 2025

Table of Contents

What's New in 2024

Learn what's new on AI Runtime Security in 2025.

Feature	Description
AI Runtime Security for Private Clouds Released in April	Introducing AI Runtime Security for private clouds. Secure your AI workloads on ESXi and KVM-based servers interacting with public cloud LLM models. To get started: Get the AI Runtime Security licenses and create a deployment profile in the Customer Support portal. Download the AI Runtime Security base image for the private cloud (ESXi or KVM). Deploy AI Runtime Security: Network intercept by manually bootstrapping the firewall on your private cloud infrastructure. AI Runtime Security: Network intercept can be deployed in Active/Passive mode or as Standalone firewall. Secure with a security policy rule and push configurations from Strata Cloud Manager or Panorama to the firewall deployed in the private cloud. Manage the firewall through Strata Cloud Manager or Panorama.
AI Runtime Security: API Intercept: Malicious Code Detection Released in March	AI Security Profile Customization AI Application Protection Added Malicious Code Detection to analyze code snippets generated by Large Language Models (LLMs) and identify potential security threats. The feature supports malicious code scanning for the following supported languages: Javascript, Python, VBScript, Powershell, Batch, Shell, and Perl. Enable this detection by updating the Create and Configure API Security Profile. For details on using the scan APIs refer to the API reference documentation.
AI Runtime Security: API Intercept: Toxic Content Detection Released in March	AI Security Profile Customization AI Model Protection: Added Toxic Content Detection in LLM model requests and responses to protect the models from generating or responding to inappropriate content. Toxic content includes references to hateful, sexual, violent, or profane themes. Malicious threat actors can easily bypass the LLM guardrails against toxic content through direct or indirect prompt injection. Enable this detection by updating the API security profile. For details on using the scan APIs refer to the API reference documentation.
AI Runtime Security: Network Intercept Managed by Panorama Released in February	You can now manage and monitor your AI Runtime Security: Network intercept (AI firewall) with Panorama. AI security policy and logs can now be defined and observed on Panorama. To get started: Select “Panorama for Management (with Log Collector)” when creating a deployment profile for Panorama in the Customer Support Portal. Generate the VM Auth Key on Panorama for AI network intercept. Deploy AI Runtime Security: Network intercept to be managed by Panorama. Export your AI network intercept logs to Panorama.
AI Runtime Security: API Intercept Released in January	Manage Applications, API Keys, and Security Profiles Create and manage multiple API keys. Define and manage multiple applications. Create and manage multiple AI security profiles and their revisions. AI Security Profile Customization AI Application Protection: Enhanced the application security with advanced options for URL filtering with custom allow and block lists for the predefined URL security categories. AI Data Protection: Expanded data loss prevention (DLP) profile selection - You can now define your custom DLP profiles for AI security. Database Security Detection: Enable database security detection to regulate database security threats in the prompt or response. This feature allows you to allow or block malicious SQL queries, preventing unauthorized actions on your database. (For detailed instructions on implementing this feature and using the scan APIs, refer the API intercept overview section).