[City, State] – [Date] – In a significant move to address the escalating complexities of enterprise artificial intelligence adoption, Palo Alto Networks today announced the general availability of its Prisma AIRS AI Gateway. This innovative solution is positioned as the "AI control plane for the enterprise," aiming to bridge the critical gap between the rapid velocity of AI innovation and the imperative for robust organizational control and security. The announcement follows closely on the heels of Palo Alto Networks’ successful acquisition of Portkey, integrating its AI innovations into the Prisma AIRS platform in a remarkable six-week timeframe.
The launch signifies a proactive response to a burgeoning challenge: the exponential growth of AI usage within organizations is rapidly outstripping existing security and governance frameworks. As businesses increasingly transition from software-driven operations to AI-orchestrated environments, a fundamental tension arises between the need for agility and the necessity for oversight. The Prisma AIRS AI Gateway is designed to resolve this tension by providing a unified architecture that enables organizations to scale AI initiatives at machine speed without compromising on enterprise-grade security and control.
The Unstoppable Tide of AI: Adoption Outpacing Governance
The imperative for an AI Gateway is underscored by stark data revealing the dramatic acceleration of AI adoption. Telemetry from Palo Alto Networks’ next-generation firewalls paints a vivid picture: Machine Code Processing (MCP) activity has surged from a mere 11% in late last year to a staggering 41.4% by mid-2026. This widespread adoption is not confined to niche applications; it is deeply embedded in core business functions.
Furthermore, the volume of AI transactions has exploded, growing twelve-fold over the same six-month period. The implications are profound, with some individual sessions observed to be moving hundreds of megabytes of sensitive enterprise data outbound. This indicates that the "AI footprint" that organizations can currently govern is, in fact, the smallest it will ever be, highlighting an urgent need for proactive management.
The most rapid AI adoption is currently being witnessed in critical areas such as:
- Coding Assistants: Empowering developers to write, debug, and optimize code at unprecedented speeds.
- Enterprise Agents: Automating complex workflows and tasks across various business units.
- Copilots: Providing intelligent assistance and augmentation for a wide range of user activities.
As this enterprise data moves seamlessly with AI requests, it introduces a security and governance challenge of an unprecedented scale and velocity. The inherent nature of AI, where information is processed and transmitted rapidly, creates a complex web of potential vulnerabilities and data leakage points that traditional security models are ill-equipped to handle.
The Fragmentation of AI Security: What Breaks When Every Team Adopts AI
The rapid, decentralized adoption of AI by individual teams and departments often leaves established security and governance infrastructure struggling to keep pace. This leads to a critical breakdown in several key areas:
- Shadow AI Infrastructure: Teams often adopt AI tools and services without the explicit knowledge or oversight of IT and security departments, leading to an unmanageable "shadow IT" landscape for AI.
- Data Exfiltration Risks: Sensitive intellectual property, customer data, and proprietary code can be inadvertently or maliciously exposed as AI models process and transmit information.
- Compliance and Regulatory Violations: The lack of centralized control and visibility makes it difficult to ensure adherence to industry regulations and data privacy laws.
- Uncontrolled Costs: Without proper monitoring and management, the consumption of AI resources can lead to significant and unexpected cost overruns, impacting financial predictability.
- Security Vulnerabilities: Unsecured AI agents and applications can become entry points for attackers, leading to data breaches and system compromises.
Faced with this complex landscape, many organizations find themselves at an impasse. Leaders are often forced to choose between two failing approaches:
- Blocking AI Traffic Entirely: This extreme measure cripples innovation and productivity, hindering the organization’s ability to leverage the transformative power of AI.
- Remaining Permissive and Promising to Govern Later: This approach creates a significant security and compliance risk, leaving the organization vulnerable to immediate threats and long-term repercussions.
Both of these strategies fail because they bypass a critical, foundational step: the ability to see what AI agents are doing at runtime and control their actions in real-time. This is where the Prisma AIRS AI Gateway emerges as a crucial solution.
Accelerating AI Adoption with Unified Control: The Prisma AIRS AI Gateway
To enable safe and scalable AI adoption, organizations require a single, unified control plane that sits at the nexus of every AI interaction and the underlying backend models. Palo Alto Networks’ Prisma AIRS AI Gateway delivers precisely this, consolidating AI governance, identity, and runtime controls into a singular, powerful platform.
With the AI Gateway, organizations can achieve the following critical objectives:
- Gain Comprehensive Visibility: Understand exactly how AI is being used across the enterprise, who is using it, and for what purpose.
- Enforce Granular Security Policies: Implement robust security measures to protect sensitive data and intellectual property from unauthorized access or exfiltration.
- Ensure Regulatory Compliance: Meet stringent compliance requirements by maintaining audit trails and enforcing data governance policies.
- Optimize AI Costs: Monitor and manage AI resource consumption to control spending and maximize return on investment.
- Empower Innovation Safely: Provide developers and business units with the freedom to explore and leverage AI technologies without introducing undue risk.
How the Prisma AIRS AI Gateway Works: A Unified Enforcement Point
The Prisma AIRS AI Gateway operates as an inline component, strategically positioned between every AI interaction, model provider, and agentic interaction. This includes connections between:
- Agents/AI Applications and Large Language Models (LLMs)
- Machine Code Processing (MCP) Tool Calls
- Application-to-Application (A2A) AI communications
Essentially, it functions as a unified gateway for LLMs, MCP, and A2A interactions, serving as a single enforcement point for all operational and security controls. This allows existing development workflows and tools, such as coding assistants, enterprise agents, and copilots, to continue functioning seamlessly, while the critical enforcement of policies shifts to the infrastructure layer, managed by dedicated platform teams.
The capabilities delivered through this unified control plane are comprehensive and designed to address the full spectrum of AI security and management needs:
Observability: Illuminating the AI Landscape
- Unified View: Every AI request is mapped to a single, unified view, providing complete visibility into usage patterns, individual users, projects, token counts, latency, and associated costs.
- Retire Shadow AI: This comprehensive observability enables organizations to immediately identify and retire any unsanctioned or "shadow" AI infrastructure, bringing all AI activity under central management.
Governance: Steering AI with Precision
- Centralized Control: Define and enforce approved AI models, tools, and access policies centrally, without requiring any modifications to developer configurations.
- FinOps and Usage Management: Track the cost, token consumption, and latency of every AI request, broken down by team or project. This provides instant answers to cost-related queries, allows for the immediate shutdown of unsanctioned AI usage, and proactively enforces budgets and rate limits before access is even granted.
Coding Assistant Security: Safeguarding Development Assets
- Credential Protection: Securely manage and protect sensitive credentials, proprietary code, and development systems from potentially risky actions taken by AI coding assistants.
- Scoped Credentials: The gateway replaces raw provider keys with scoped credentials that are specific to each user and team, minimizing the blast radius of any potential compromise.
Operational Controls: Real-time Policy Enforcement
- Data Protection: Apply data protection policies, usage limits, and critical checks as AI interactions occur in real-time.
- Universal API Distribution: Traffic is distributed across providers via a Universal API, ensuring that quotas are consistently enforced and that potential outages do not disrupt critical AI pipelines.
Agent Identity Security: Establishing Trust in AI Interactions
- Verifiable Identities: Establish trusted identities by binding a verifiable, ephemeral identity to AI agents at the point of execution.
- Authenticated Access: The AI Gateway acts as an enforcement point, ensuring that only authenticated agents making approved calls are granted access, thereby preventing unauthorized or malicious agent activity.
Runtime Security: Proactive Threat Neutralization
- Inline Inspection: Powered by Prisma AIRS AI Runtime Security, the gateway inspects every prompt and response inline.
- Data Leakage Prevention: This real-time inspection prevents the exfiltration of sensitive information such as source code, secrets, and customer data from leaving the network.
- Prompt Injection Defense: The system effectively neutralizes prompt injection attempts, a critical threat vector, by aligning with the latest security best practices, including the OWASP LLM Top 10 and the OWASP Top 10 for Agentic Applications.
A Foundation for the Agentic Enterprise
The Prisma AIRS AI Gateway is more than just a product; it represents a fundamental shift in how enterprises approach AI security and management. It brings together AI governance, identity, and runtime controls into a single, cohesive platform. While point solutions may offer superficial filtering of text strings or the routing of individual API calls, they often falter under the immense volume and complexity of enterprise-grade AI workloads.
The Prisma AIRS AI Gateway is built upon an architecture that has been rigorously tested and proven in the most demanding enterprise environments, capable of adapting to the ever-evolving landscape of AI workloads. This robust foundation ensures that organizations can confidently embrace AI without sacrificing security or control.
Palo Alto Networks has been recognized by Gartner as the "company to beat" in AI Security Platforms, a testament to its deep understanding of the evolving threat landscape and its commitment to delivering innovative solutions. The company is uniquely positioned to guide organizations through the transition to the agentic enterprise, ensuring that this powerful new era of computing is built on a secure and controlled foundation.
Organizations are invited to learn more about the Prisma AIRS AI Gateway and register for an upcoming webinar to witness its capabilities in action. As the agentic enterprise becomes a reality, Palo Alto Networks stands ready to partner with businesses to build and scale their AI initiatives securely, together.
