Palo Alto Networks, a worldwide cybersecurity leader, has just rolled out a major extension of its AI security features through the launch of Prisma AIRS 3. 0. Modern enterprises are moving toward autonomous systems, and this new version is intended to provide end-to-end security of the Agentic AI lifecycle. It allows organizations to not only monitor AI interactions passively but also to carry out autonomous activities securely and with proper authorizations.
The emergence of the AI-driven enterprise has brought about complex security hurdles, including the rise of unmanaged “Shadow AI,” risks associated with agentic identity, and the need for automated governance during runtime. While many current security frameworks focus on monitoring AI outputs, they often lack visibility into AI actions. Prisma AIRS 3.0 is designed to bridge this gap, offering comprehensive oversight and protection for autonomous agents from the initial design phase through real-time execution.
Anand Oswal, Executive Vice President of AI & Network Security at Palo Alto Networks, commented on the shift: “Agentic AI represents a massive leap forward, moving beyond simple conversation to autonomous action that will redefine productivity. But this shift from ‘AI that talks’ to ‘AI that acts’ introduces new risks – from unmanaged agentic identities to unpredictable runtime behaviors. Prisma AIRS 3.0 provides a comprehensive platform to discover, assess and protect agentic AI, giving our customers the unique ability to confidently, and securely, scale the AI-powered enterprise.”
Also Read: Zluri Unveils Enhanced Identity Security Platform to Safeguard the Modern Enterprise Ecosystem
By consolidating fragmented security tools into a unified platform, Prisma AIRS 3.0 addresses the primary threats facing AI applications and autonomous agents.
The updated suite introduces several key features to help organizations future-proof their operations:
- Comprehensive AI Agent Discovery: This platform offers an immediate and comprehensive list of all AI models, agents, and their connections throughout the entire enterprise. This platform is successful in discovering agents in SaaS platforms, cloud, and endpoints, which often evade traditional security solutions.
- Continuous Risk Assessment: Security teams can now utilize “Agent Artifact Security” to map out architectures and scan for potential vulnerabilities. This platform also utilizes “AI Red Teaming” to carry out sophisticated and contextualized attacks, revealing potential vulnerabilities and providing data-driven recommendations for security policies.
- Real-Time Ecosystem Protection: This new “AI Agent Gateway” is currently in limited preview and acts as a control plane to manage and govern both agent identity and runtime. Furthermore, with the acquisition of Koi, “Agentic Endpoint Security” will offer the required visibility to deploy high-speed AI tools such as coding agents, while still ensuring safety.
Harpreet Sidhu, Global Lead for Accenture Cybersecurity, highlighted the importance of these advancements: “Scaling agentic AI requires clearing critical hurdles around visibility and control. To do this, organizations must know where agents reside, what data they access and how they behave post-deployment. By providing a comprehensive, end-to-end view of the AI lifecycle, Prisma AIRS helps clients mitigate risk. We are excited to collaborate with Palo Alto Networks to help enterprises enforce consistent guardrails allowing them to scale agentic AI securely.”
