Palo Alto Networks launches Prisma AIRS 2.0, a platform upgrade that provides businesses with comprehensive security for AI applications, from autonomous agents to underlying models. This expansion follows the integration of Protect AI.
Palo Alto Networks introduces Prisma AIRS 2.0, which focuses on improved AI security. According to the company, 78 percent of organizations use AI, but only six percent have adequate security measures in place. The platform provides protection across the entire AI lifecycle. It secures interactions between models, agents, data, and users, both during development and in production.
Three Security Modules
The update consists of three security modules. AI Agent Security detects and secures AI agents, even when they operate outside IT department visibility (Shadow AI). The module defends against prompt injections, tool misuse, and unwanted AI agent behavior.
read also
Palo Alto Acquires Protect AI and Sets its Sights on SentinelOne
AI Red Teaming performs autonomous and continuous attacks to identify vulnerabilities. The system uses more than 500 attack techniques to proactively test AI applications, unlike traditional, periodic testing.
AI Model Security analyzes AI model architecture in depth. It detects threats that regular scanners miss, such as hidden backdoors or data poisoning. This module also provides a complete overview of model structure, datasets used, software dependencies, and licenses, which is important for governance and compliance.
read also
Palo Alto Networks Introduces Prisma SASE 4.0 with Browser Security
Prisma AIRS 2.0 is available now. The solution targets sectors such as finance, healthcare, and government. With this expansion, Palo Alto Networks aims to help businesses implement AI in a controlled and secure manner.