ProtectAI

This listing now resolves to Prisma AIRS, the AI security platform from Palo Alto Networks, which is where the Protect AI technology landed after the acquisition. The platform is designed to discover, assess and protect an organization's entire AI estate from one console, covering agents, applications, models and the data behind them from development through to production. It targets enterprise security teams whose organizations have moved from AI pilots to broad deployment. Discovery maps every AI agent, application and model in the environment, including shadow AI, and shows how they connect. Assessment continuously tests applications and agents, controls permissions and reports posture in real time. Protection enforces AI specific controls at runtime. The platform is organized into modules: AI Agent Security verifies agent identity and blocks unauthorized actions as deployments scale; AI Red Teaming runs automated penetration tests against your own applications and models, adapting as a real attacker would; AI Runtime Security defends language model applications against prompt injection, malicious code, toxic content, sensitive data leakage, resource overload and hallucination; AI Model Security scans third party models for tampering, malicious scripts and deserialization attacks; and AI Posture Management covers training and inference data, model integrity and access to deployed models. A fully managed AI runtime security service, delivered as a hosted firewall in front of containers, applications and cloud infrastructure, is available in early access. Pricing is not published. Palo Alto Networks lists no amounts for Prisma AIRS and routes buyers through a demonstration request and its sales organization. The rename matters when looking for documentation or release notes, since all current material is filed under Prisma AIRS rather than the original brand.