Palo Alto Networks has strengthened its AI security platform with the launch of Prisma AIRS 3.0. As organizations move toward a future marked by AI and autonomous agents, Prisma AIRS 3.0 protects the entire agentic AI lifecycle, enabling enterprises to move from simply observing AI interactions to securely authorizing AI-powered autonomous executions.
The transition to an AI-driven enterprise raises systemic security challenges, ranging from unmanaged shadow AI to critical new frontiers such as AI agent identity, AI runtime security, and automated AI governance. Although many companies monitor what AI says, they still don’t know what it does. Prisma AIRS 3.0 closes this gap, providing visibility and protecting AI agents from design to execution, as they autonomously carry out complex AI-based tasks.
According to Anand Oswal, executive vice president of AI and Network Security at Palo Alto Networks: “Agentic AI is a significant advance, moving beyond simple conversation to autonomous AI-based action, which will redefine productivity. However, this shift from AI that talks to AI that acts brings new risks, from unmanaged agentic AI identities to unpredictable AI runtime behaviors. Prisma AIRS 3.0 provides a platform comprehensive to detect, evaluate and protect agentic AI, enabling our customers to confidently and securely scale their AI-powered business.”
Evolution of agent ecosystems
Prisma AIRS replaces fragmented point solutions with a single platform that manages the main threats and risks of AI applications and autonomous AI agents. New capabilities enable teams to future-proof their operations as AI ecosystems evolve:
● Discover AI agents wherever they are. Organizations can now take an instant inventory of AI agents, models, and connections across their environment. Prisma AIRS identifies AI agents running in cloud environments, SaaS platforms, and on-premises at endpoints that traditional tools miss.
● Continuous risk assessment of AI agents. Security teams will no longer have to guess whether an AI agent is safe. Agent Artifact Security analyzes the architecture of an AI agent and looks for vulnerabilities. AI Red Teaming for agents simulates specific attacks against AI systems, detects AI-related vulnerabilities, and recommends runtime security policies for AI environments.
● Protecting AI ecosystems in real time and at scale. The AI Agent Gateway, currently available in limited preview, provides a centralized control plane to ensure the identity and runtime security of AI agents, as well as AI governance and observability. Upon completion of the proposed acquisition of Koi, Agentic Endpoint Security will provide the visibility needed to protect various AI applications at endpoints, allowing teams to deploy tools such as AI-based encryption agents quickly and without sacrificing security.
Prisma AIRS 3.0 offers a comprehensive platform to detect, evaluate and protect agentic AI
According to Harpreet Sidhu, Global Leader, Accenture Cybersecurity: “Scaling agentic AI requires overcoming fundamental barriers to AI visibility and control. To do so, organizations must understand where AI agents are located, what data they access, and how they behave after deployment. By providing a comprehensive, comprehensive view of the AI lifecycle, Prisma AIRS helps clients mitigate the risks associated with AI. We are proud to partner with Palo Alto Networks to help companies establish consistent security measures that allow them to safely scale agentic AI.”
