Palo Alto, CA – [Insert Date] – In a significant move to address the burgeoning challenges of enterprise artificial intelligence adoption, Palo Alto Networks today announced the general availability of its Prisma AIRS AI Gateway. Positioned as the "AI control plane for the enterprise," this groundbreaking solution aims to bridge the critical gap between the rapid velocity of AI innovation and the imperative for robust security and control. The launch underscores Palo Alto Networks’ commitment to enabling organizations to harness the transformative power of AI at machine speed, without compromising on enterprise-grade security.
The announcement arrives just six weeks after the successful completion of Palo Alto Networks’ acquisition of Portkey, a testament to the company’s agile integration of cutting-edge AI capabilities. This rapid infusion of Portkey’s innovations into Prisma AIRS signals a proactive approach to the evolving AI landscape, acknowledging that the future of enterprise operations is intrinsically linked to intelligent automation.
The Unprecedented Acceleration of AI Adoption: A Growing Control Deficit
The modern enterprise is undergoing a profound transformation, shifting from a software-driven operational model to one orchestrated by artificial intelligence. This seismic shift, while promising unparalleled efficiency and innovation, has also introduced a fundamental tension between the need for rapid deployment and the necessity for stringent control. Palo Alto Networks’ internal telemetry paints a stark picture of this accelerating trend.
Data from Palo Alto Networks’ next-generation firewalls reveals a dramatic surge in the adoption and utilization of Machine Control Plane (MCP) activities. Late last year, these activities constituted a mere 11% of enterprise traffic. By mid-2026, this figure has skyrocketed to an astonishing 41.4%. This exponential growth is mirrored in the sheer volume of AI transactions. Over a six-month period, monthly AI transaction volume has grown twelve-fold, with some individual sessions exhibiting concerning patterns of outbound data transfer, moving hundreds of megabytes of sensitive enterprise information.
This rapid proliferation of AI, particularly in critical areas such as coding assistants, enterprise agents, and copilots, is fundamentally outpacing the traditional security and governance frameworks designed to manage them. As these intelligent tools interact with external models and services, enterprise data – including proprietary code, sensitive credentials, and confidential customer information – is being exposed at an unprecedented scale and speed. This creates a complex and rapidly evolving security and governance challenge that existing infrastructures are ill-equipped to handle.
The Cracks Appear: When Every Team Embroys AI
The widespread adoption of AI across various departments within an organization, while a driver of innovation, simultaneously exposes vulnerabilities in the existing security and governance infrastructure. The core issue lies in the inherent lag between the pace of AI development and the deployment of corresponding security measures.
Organizations are increasingly finding themselves at a critical juncture: either completely block AI-driven traffic, stifling innovation and productivity, or adopt a permissive stance, promising to implement controls at a later, unspecified date. Both approaches are inherently flawed. Blocking AI traffic can lead to a significant competitive disadvantage and hinder employee efficiency. Conversely, a "govern later" strategy leaves organizations exposed to immediate and potentially catastrophic security breaches and compliance violations. The critical step of real-time visibility into AI agent actions and the ability to control their behavior as they happen is often bypassed, leaving a dangerous void in enterprise security.
This dilemma arises because AI adoption is not a monolithic, centrally managed process. Instead, it is often a decentralized movement, with individual teams independently integrating AI tools and agents to enhance their workflows. Without a unified oversight mechanism, this distributed adoption creates a fragmented and largely ungoverned AI footprint, making it exceedingly difficult to maintain visibility, enforce policies, and mitigate risks.
Accelerating AI Adoption with Unified Control: The Prisma AIRS AI Gateway Solution
To navigate this complex landscape and enable safe, scalable AI adoption, organizations require a singular, authoritative control plane. This crucial layer must sit strategically between every AI interaction and the backend models, acting as a central point of governance, identity management, and runtime control. Palo Alto Networks’ Prisma AIRS AI Gateway is engineered to fulfill this critical need, consolidating these vital functions into a single, unified platform.
The AI Gateway empowers organizations to:
- Gain Unprecedented Visibility: Obtain a comprehensive, unified view of all AI activity, tracking usage patterns, individual users, specific projects, token consumption, latency, and associated costs. This eliminates "shadow AI" infrastructure and provides the data needed for informed decision-making.
- Implement Robust Governance: Centrally define and enforce approved AI models, tools, and access policies without requiring developers to alter their configurations. This granular control facilitates effective FinOps, enabling precise tracking of costs, token usage, and latency per team or project. Organizations can proactively identify and shut down unsanctioned AI usage, enforce budgets, and implement rate limits before unauthorized access is granted.
- Secure Coding Assistants: Safeguard critical assets such as credentials, proprietary code, and development systems from the inherent risks associated with AI actions. The AI Gateway achieves this by replacing raw provider keys with secure, scoped credentials tailored to individual users and teams.
- Enforce Operational Controls: Apply essential data protection measures, usage limits, and policy checks in real-time as AI interactions occur. A Universal API ensures traffic distribution across providers, guaranteeing quota enforcement and preventing pipeline disruptions due to provider outages.
- Establish Agent Identity Security: Institute trusted identities by binding verifiable, ephemeral identities to agents at the point of execution. The AI Gateway acts as a critical enforcement point, ensuring that only authenticated agents can perform approved calls, thereby mitigating the risk of unauthorized or malicious agent behavior.
- Strengthen Runtime Security: Leverage the advanced capabilities of Prisma AIRS AI Runtime Security to inspect every prompt and response inline. This ensures that sensitive information like source code, secrets, and customer data are prevented from leaving the network. Furthermore, it actively neutralizes threats such as prompt injection, aligning with the critical security guidelines outlined in the OWASP LLM Top 10 and the OWASP Top 10 for Agentic Applications.
The Architecture of Control: How the AI Gateway Operates
The Prisma AIRS AI Gateway is designed to function as an inline security and control layer for all AI interactions. It sits between users, AI applications, and backend Large Language Models (LLMs), acting as a unified gateway for LLMs, Machine Control Plane (MCP) tools, and Agent-to-Agent (A2A) communications. This singular enforcement point ensures that all operational and security controls are applied consistently and effectively.
Developers can continue to utilize their preferred coding assistants, enterprise agents, and copilots without disruption. The enforcement of security and governance policies is seamlessly shifted to the infrastructure layer, where platform teams can maintain centralized ownership and oversight.
The capabilities delivered through this unified control plane are multifaceted and address the core challenges of enterprise AI adoption:
Observability: Illuminating the AI Landscape
The Prisma AIRS AI Gateway provides a single, unified view of every AI request. This comprehensive observability allows organizations to meticulously track usage, identify individual users, monitor project-specific AI consumption, quantify token counts, measure latency, and analyze associated costs. This granular insight is crucial for retiring redundant or "shadow AI" infrastructure and for making data-driven decisions about AI resource allocation and optimization.
Governance: Establishing the Rules of Engagement
At its core, the AI Gateway is a powerful governance engine. It allows for the central definition and enforcement of approved models, tools, and access permissions, eliminating the need for developers to manage these configurations directly. This is particularly critical for driving effective FinOps and usage management. By tracking every request’s cost, token usage, and latency segmented by team or project, organizations can instantly address cost-related inquiries, swiftly curtail unsanctioned AI usage, and proactively enforce budgets and rate limits before access is even granted.
Coding Assistant Security: Protecting Development Lifelines
The integration of AI into software development workflows, while boosting productivity, introduces new risks. The Prisma AIRS AI Gateway provides robust protection for credentials, proprietary code, and development systems against potentially risky AI actions. It achieves this by replacing raw provider API keys with securely scoped credentials that are specific to each user and team, significantly reducing the attack surface.
Operational Controls: Real-time Policy Enforcement
The gateway ensures that critical operational controls are applied dynamically as AI interactions unfold. This includes robust data protection measures, the enforcement of usage limits, and comprehensive policy checks. The Universal API architecture facilitates seamless traffic distribution across various AI providers, guaranteeing that quotas are strictly enforced and that potential provider outages do not disrupt critical AI pipelines.
Agent Identity Security: Verifying Trust in AI Agents
In an agentic enterprise, establishing trusted identities for AI agents is paramount. The AI Gateway enables this by binding a verifiable, ephemeral identity to agents at the point of execution. This ensures that only authenticated and authorized agents can initiate approved calls, acting as a crucial enforcement point to prevent rogue or compromised agents from operating within the enterprise environment.
Runtime Security: Defending Against Evolving Threats
Powered by Prisma AIRS AI Runtime Security, the gateway meticulously inspects every prompt and response in real-time. This inline inspection capability is vital for preventing the exfiltration of sensitive data, including source code, secrets, and customer information. Furthermore, it actively neutralizes sophisticated threats like prompt injection, aligning with the latest security best practices outlined by the OWASP LLM Top 10 and the OWASP Top 10 for Agentic Applications. This proactive defense mechanism is essential for maintaining the integrity and security of AI-driven operations.
The Foundational Role of an AI Gateway
Anand Oswal, Executive Vice President and General Manager of Prisma Cloud at Palo Alto Networks, emphasized the critical nature of an AI gateway in modern enterprise architecture. "Every new AI agent creates another decision point, another potential vulnerability," Oswal stated. "Without a unified control plane, organizations are left with a fragmented and unmanageable AI footprint. Prisma AIRS AI Gateway is not just a product; it’s a foundational element for securing the agentic enterprise, enabling innovation at scale while maintaining absolute control."
A Proven Architecture for the Demanding Enterprise
The Prisma AIRS AI Gateway is built upon an architecture that has been rigorously tested and proven in the most demanding enterprise environments. This robust foundation is designed to support the ever-evolving and increasingly complex AI workloads that characterize modern businesses. Unlike point solutions that often struggle to scale or offer comprehensive protection, Prisma AIRS AI Gateway provides a unified and resilient platform for managing the entire AI lifecycle.
Palo Alto Networks’ leadership in AI security, recognized by Gartner as the "company to beat" in AI Security Platforms, positions it uniquely to guide organizations through this transformative period. The launch of Prisma AIRS AI Gateway represents a significant step forward in enabling enterprises to build and operate AI securely and confidently.
Organizations are encouraged 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 remains committed to partnering with businesses to build a secure AI future, together.
