[City, State] – [Date] – In a pivotal moment for enterprise digital transformation, Palo Alto Networks today announced the general availability of Prisma AIRS AI Gateway, a comprehensive AI control plane designed to address the escalating tension between the rapid velocity of AI adoption and the imperative for robust security and control. This groundbreaking solution, integrating innovations from the recent Portkey acquisition, aims to empower organizations to harness the full potential of AI at machine speed without compromising their critical security and governance frameworks.
The enterprise landscape is undergoing a profound metamorphosis. No longer are businesses simply run by software; they are increasingly orchestrated by artificial intelligence. This seismic shift, while promising unprecedented efficiency and innovation, introduces a complex dichotomy: the drive for speed and agility versus the fundamental need for oversight and security. Palo Alto Networks posits that a unified architecture is not merely beneficial, but foundational to navigating this new paradigm.
The Accelerating AI Footprint: A Growing Challenge for Enterprise Control
The statistics paint a stark picture of AI’s burgeoning influence within modern enterprises. Telemetry data from Palo Alto Networks’ next-generation firewalls reveals a dramatic surge in Machine Conversation Platform (MCP) activity, climbing from a mere 11% late last year to a significant 41.4% by mid-2026. This surge is mirrored by an astounding twelve-fold increase in monthly AI transaction volume over the same six-month period. More alarmingly, some individual AI sessions have been observed exfiltrating hundreds of megabytes of sensitive enterprise data. This rapid expansion of the "AI footprint" means that the scope of what organizations can govern today is the smallest it will ever be.
The most aggressive AI adoption is currently concentrated in areas such as coding assistants, enterprise agents, and copilots. These powerful tools, while instrumental in boosting productivity, inherently involve the transmission of enterprise data beyond traditional security perimeters with every outbound request. This presents a security and governance challenge of an unprecedented scale and velocity, demanding a new class of solutions to manage the inherent risks.
The Breakdown of Traditional Controls in the Face of Pervasive AI
As every team within an organization begins to embrace AI, the existing security and governance infrastructure often struggles to keep pace. This disconnect can manifest in several critical ways:
- Shadow AI Proliferation: Without centralized visibility and control, teams may independently adopt AI tools and services, leading to "shadow AI" environments that are unknown to IT and security teams. This creates blind spots, making it impossible to monitor usage, enforce policies, or manage costs effectively.
- Data Leakage Risks: The seamless integration of AI agents with enterprise data sources can inadvertently expose sensitive information. Without proper safeguards, proprietary code, customer data, and intellectual property can be shared with AI models or external services without explicit consent or oversight.
- Inconsistent Security Posture: A fragmented approach to AI adoption results in a patchwork of security controls, leaving gaps that can be exploited by malicious actors. Different teams may implement varying levels of security, leading to an overall weakened defense.
- Uncontrolled Costs: The consumption of AI services, particularly through large language models (LLMs), can incur significant costs. Without mechanisms to track usage, enforce budgets, and optimize resource allocation, organizations risk runaway AI spending.
- Compliance and Governance Gaps: The dynamic nature of AI interactions makes it challenging to ensure compliance with regulatory requirements and internal governance policies. Auditing AI usage and demonstrating adherence to these standards becomes an arduous task.
- Prompt Injection Vulnerabilities: AI models are susceptible to "prompt injection" attacks, where malicious prompts can manipulate the AI’s behavior, leading to unintended actions or the leakage of sensitive information.
Faced with these escalating challenges, many leaders find themselves at an impasse. The two common, yet ultimately flawed, responses are to either block AI traffic entirely, stifling innovation and productivity, or to remain permissive, promising to implement governance measures later. Both approaches fail to address the critical need to see what AI agents are doing in real-time and to control their actions as they occur.
Prisma AIRS AI Gateway: Unifying Velocity and Control for the Agentic Enterprise
To effectively scale AI adoption while mitigating these risks, organizations require a single, unified control plane that sits at the nexus of every AI interaction and the underlying AI models. Palo Alto Networks’ Prisma AIRS AI Gateway is precisely this solution, bringing together AI governance, identity, and runtime controls into a singular, powerful platform.
The general availability of Prisma AIRS AI Gateway marks a significant milestone, fueled by the strategic acquisition of Portkey, whose AI innovations have been seamlessly integrated into the Prisma AIRS platform in an unprecedented six weeks. This rapid integration underscores Palo Alto Networks’ commitment to delivering cutting-edge AI security solutions.
With Prisma AIRS AI Gateway, organizations can:
- Gain Unprecedented Visibility: Achieve a unified view of all AI activity, tracking usage, users, projects, token counts, latency, and costs to swiftly identify and retire shadow AI infrastructure.
- Enforce Granular Governance: Centrally define and enforce approved AI models, tools, and access policies without altering developer configurations. Monitor and manage costs per team or project, enabling proactive budget enforcement and rate limiting.
- Secure Coding Assistants: Protect sensitive credentials, proprietary code, and development systems by replacing raw provider keys with scoped, per-user, and per-team credentials.
- Implement Robust Operational Controls: Apply data protection measures, usage limits, and policy checks in real-time as AI interactions occur. A Universal API ensures traffic distribution across providers, enforcing quotas and preventing outages from disrupting critical AI pipelines.
- Establish Trusted Agent Identities: Bind verifiable, ephemeral identities to AI agents at execution, ensuring that only authenticated and approved agents can make authorized calls.
- Deliver Advanced Runtime Security: Inspect every prompt and response inline to prevent the leakage of sensitive data, neutralize prompt injection attempts, and align with the latest security frameworks like the OWASP LLM Top 10 and the OWASP Top 10 for Agentic Applications.
How the Prisma AIRS AI Gateway Operates: A Centralized Enforcement Point
The Prisma AIRS AI Gateway functions as an inline intermediary for every AI interaction. It sits between users, AI applications, and backend AI models, acting as a unified gateway for Large Language Models (LLMs), Machine Conversation Platforms (MCPs), and Agent-to-Agent (A2A) communications. This single enforcement point consolidates all operational and security controls, allowing development teams to continue utilizing their preferred coding assistants, enterprise agents, and copilots while the platform team maintains centralized oversight and control at the infrastructure layer.
The capabilities delivered through this unified control plane are comprehensive and designed for the demands of the agentic enterprise:
Observability: Illuminating the AI Landscape
In the race to adopt AI, visibility has become a critical casualty. The Prisma AIRS AI Gateway offers a single, unified view of all AI requests, meticulously tracking usage patterns, individual users, project allocations, token consumption, latency metrics, and associated costs. This level of granular insight is essential for identifying and eliminating "shadow AI" – unauthorized or unmanaged AI applications – and for optimizing resource utilization. By retiring redundant or inefficient AI infrastructure, organizations can achieve significant cost savings and streamline their AI operations.
Governance: Steering AI with Precision
Establishing effective governance is paramount for responsible AI deployment. The Prisma AIRS AI Gateway enables centralized definition and enforcement of approved AI models, tools, and access policies, eliminating the need for individual developers to manage these configurations. This empowers platform teams to maintain a consistent security posture across the organization. Furthermore, the gateway facilitates robust FinOps and usage management by precisely tracking the cost, token usage, and latency of every AI request, categorized by team or project. This allows for instant answers to cost-related queries, the immediate shutdown of unsanctioned AI usage, and proactive enforcement of budgets and rate limits before access is granted, preventing overspending and ensuring compliance.
Coding Assistant Security: Fortifying Development Workflows
Coding assistants, while invaluable for developer productivity, can also pose significant risks if not properly secured. The Prisma AIRS AI Gateway acts as a critical safeguard, protecting sensitive credentials, proprietary code, and development systems from potentially risky AI actions. By replacing raw provider keys with scoped, ephemeral credentials that are specific to each user and team, the gateway significantly reduces the attack surface and prevents unauthorized access to sensitive resources. This ensures that the power of AI-driven coding assistance can be harnessed without compromising the security of the development environment.
Operational Controls: Ensuring Seamless and Secure AI Execution
The dynamic nature of AI interactions necessitates sophisticated operational controls. The Prisma AIRS AI Gateway applies essential data protection measures, usage limits, and policy checks in real-time as AI interactions unfold. Its Universal API ensures that traffic is efficiently distributed across various AI providers, guaranteeing that quotas are consistently enforced and that potential outages with one provider do not stall the entire AI pipeline. This resilience and control are vital for maintaining business continuity and ensuring the reliable operation of AI-powered applications.
Agent Identity Security: Establishing Trust in the AI Ecosystem
As AI agents become more sophisticated and autonomous, establishing their identity and ensuring their trustworthiness is crucial. The Prisma AIRS AI Gateway introduces Agent Identity Security, which binds a verifiable, ephemeral identity to agents at the point of execution. This acts as a robust enforcement mechanism, allowing only authenticated agents to make approved calls to AI models and services. This feature is fundamental to building a secure and reliable agentic enterprise, where each AI actor can be reliably identified and its actions audited.
Runtime Security: Proactive Defense Against Evolving Threats
Leveraging the power of Prisma AIRS AI Runtime Security, the AI Gateway provides deep inline inspection of every AI prompt and response. This sophisticated capability is designed to prevent the exfiltration of sensitive data, including source code, secrets, and customer information, from leaving the network. Moreover, it actively neutralizes prompt injection attempts, a prevalent threat vector, by aligning with the comprehensive security guidelines of both the OWASP LLM Top 10 and the OWASP Top 10 for Agentic Applications. This proactive defense mechanism ensures that AI interactions are not only functional but also secure against emerging threats.
The Foundation for the Agentic Enterprise
The agentic enterprise is no longer a distant vision; it is a present reality. As developers continue to embrace AI agents, the need for a unified and comprehensive control plane becomes increasingly critical. Palo Alto Networks, recognized by Gartner as a leading innovator in AI Security Platforms, is uniquely positioned to guide organizations through this transformative period.
The Prisma AIRS AI Gateway is built on an architecture that has been rigorously tested and proven in the most demanding enterprise environments, designed to adapt to the ever-evolving landscape of AI workloads. Unlike point solutions that may struggle to handle the scale and complexity of enterprise AI, the Prisma AIRS AI Gateway offers a unified, robust, and scalable solution.
Organizations are invited to learn more about Prisma AIRS AI Gateway and register for an upcoming webinar to witness its capabilities in action. As the recently named "company to beat" in AI Security Platforms by Gartner, Palo Alto Networks is committed to collaborating with enterprises to build a secure and prosperous AI-driven future.
"Every new AI agent creates another decision point for enterprises," states Anand Oswal, Senior Vice President of Prisma Cloud at Palo Alto Networks. "The AI Gateway is foundational to this architecture, providing the necessary control and visibility to scale AI safely and effectively."
About Palo Alto Networks
Palo Alto Networks is the global cybersecurity leader, building a world where we can safely harness the technology of today and tomorrow. Our knowledge and tools empower organizations to secure everything: network, cloud, devices, and applications. We are at the forefront of innovation, delivering cutting-edge solutions that address the most complex security challenges faced by enterprises worldwide.
Contact:
[Media Contact Name]
[Media Contact Title]
[Media Contact Email]
[Media Contact Phone Number]
