SANTA CLARA, CA – July 25, 2026 – In a move poised to redefine enterprise AI adoption, Palo Alto Networks today announced the general availability of Prisma AIRS AI Gateway. This groundbreaking solution is positioned as the definitive AI control plane for modern organizations, aiming to bridge the critical gap between the accelerating pace of AI innovation and the imperative for robust security and governance. The launch signifies a rapid integration of AI capabilities acquired through the recent Portkey acquisition, demonstrating a commitment to delivering enterprise-grade AI security solutions at machine speed.
The imperative for such a platform stems from a stark reality: enterprises are no longer simply running on software; they are increasingly being orchestrated by artificial intelligence. This fundamental shift, while promising unprecedented velocity and efficiency, introduces a significant tension with the established need for control and security. Prisma AIRS AI Gateway is designed to resolve this tension by providing a unified architecture that governs the burgeoning AI footprint within organizations.
The Unstoppable Tide of AI Adoption: Velocity Outpacing Control
The adoption of AI within enterprises has reached an inflection point, with its footprint expanding at an exponential rate. Telemetry data from Palo Alto Networks’ next-generation firewalls paints a compelling picture: "MCP activity" – a metric encompassing the use of various AI-powered tools and services – has surged from a mere 11% late last year to a staggering 41.4% by mid-2026. This dramatic increase is mirrored by an equivalent explosion in AI transaction volume. Over the same six-month period, monthly AI transaction volume has grown twelve-fold, with some individual sessions exhibiting the alarming transfer of hundreds of megabytes of sensitive enterprise data outbound. This data suggests that the AI footprint organizations can effectively govern today is likely the smallest it will ever be.
The most rapid adoption is currently being observed in areas such as coding assistants, enterprise agents, and copilots. These tools are becoming indispensable for developer productivity and business process automation. However, as enterprise data is intrinsically linked to AI requests, its egress with every AI interaction creates a security and governance challenge of unprecedented scale and speed.
The Cracks in Existing Frameworks: What Breaks When Every Team Embroys AI
The rapid proliferation of AI tools across every team within an organization is exposing the limitations of traditional security and governance infrastructures. These frameworks, designed for a pre-AI era, are struggling to keep pace with the dynamic and often opaque nature of AI interactions.
The consequences of this disconnect are manifold:
- Shadow AI Proliferation: Without centralized visibility and control, individual teams often adopt their own AI tools and services, leading to "shadow AI." This creates blind spots for IT and security teams, making it impossible to track data flow, manage costs, or enforce security policies.
- Data Leakage Risks: The seamless integration of AI assistants with internal systems and data repositories significantly increases the risk of sensitive information, including proprietary code, intellectual property, and customer data, being inadvertently exposed or exfiltrated.
- Security Vulnerabilities: The increasing reliance on AI models and agents introduces new attack vectors, such as prompt injection, data poisoning, and model manipulation. Existing security controls are often ill-equipped to detect and mitigate these novel threats.
- Compliance and Governance Gaps: The lack of centralized oversight makes it challenging to ensure compliance with data privacy regulations (e.g., GDPR, CCPA) and internal governance policies. Understanding who is using what AI, for what purpose, and with what data becomes an insurmountable task.
- Uncontrolled Costs: The pay-as-you-go nature of many AI services, coupled with the lack of visibility into usage patterns, can lead to runaway expenses and a significant drain on IT budgets.
Faced with this complex landscape, many organizations find themselves at an impasse. Leaders are often forced to choose between two untenable extremes: either block AI traffic entirely, stifling innovation and productivity, or remain permissive and promise to implement governance later, a strategy that leaves them vulnerable to immediate risks. Both approaches ultimately fail because they bypass the critical step of actively observing AI actions at runtime and controlling their behavior in real-time.
A Unified Vision: Accelerating AI Adoption with Unwavering Control
To safely and effectively scale AI adoption, enterprises require a single, authoritative control plane that sits strategically between every AI interaction and the underlying backend models. Prisma AIRS AI Gateway delivers precisely this by consolidating AI governance, identity, and runtime controls into a unified platform.
"The modern enterprise is rapidly evolving from an organization run by software to one orchestrated by AI," stated [Executive Name and Title, e.g., Nir Zuk, Founder and CEO of Palo Alto Networks]. "This transition creates a fundamental tension between the velocity that AI promises and the control that enterprises demand. Prisma AIRS AI Gateway is our definitive answer to this challenge, providing the AI control plane that is foundational to the modern AI infrastructure stack."
The accelerated integration of Portkey’s innovative AI security capabilities into Prisma AIRS, just six weeks after the acquisition’s closure, underscores Palo Alto Networks’ commitment to leading the AI security revolution. This rapid deployment allows organizations to now scale AI at machine speed without compromising on enterprise-grade security and control.
How the Prisma AIRS AI Gateway Works: The Central Nervous System of Enterprise AI
Prisma AIRS AI Gateway operates as an inline enforcement point, strategically positioned between every AI interaction, model provider, and agentic workflow. It functions as a unified gateway for Large Language Models (LLMs), Machine Code Platforms (MCPs), and Agent-to-Agent (A2A) communications, providing a single point of enforcement for all operational and security controls.
This architecture empowers platform teams to maintain ownership and oversight of AI usage across the organization, while developers can continue to leverage their preferred coding assistants, enterprise agents, and copilots seamlessly. The enforcement of policies and security measures is moved to the infrastructure layer, ensuring consistency and scalability.
The capabilities delivered through this unified control plane are comprehensive and transformative:
H2: Granular Observability for Comprehensive AI Insight
Prisma AIRS AI Gateway provides a single, unified view of all AI activity. This allows organizations to meticulously track usage across users, projects, and applications. Detailed metrics on token counts, latency, and associated costs provide unparalleled visibility, enabling the immediate retirement of "shadow AI" infrastructure and the accurate attribution of AI resource consumption. This deep observability is the first step towards effective management and optimization.
H2: Robust Governance for Policy Enforcement and FinOps
Centralized governance is a cornerstone of Prisma AIRS AI Gateway. Organizations can define and enforce approved models, tools, and access policies without requiring any changes to developer configurations. This empowers IT and security teams to maintain a secure AI environment while fostering innovation.
For financial operations (FinOps) and usage management, the gateway meticulously tracks the cost, token consumption, and latency of every AI request, broken down by team or project. This allows for instant answers to cost-related questions, the ability to shut down unsanctioned AI usage, and the proactive enforcement of budgets and rate limits before access is even granted. This proactive approach to cost control is essential in an era of escalating AI expenditure.
H2: Fortified Coding Assistant Security
Coding assistants are powerful tools, but they also pose risks to critical assets such as credentials, proprietary code, and development systems. Prisma AIRS AI Gateway addresses this by replacing raw provider keys with scoped credentials that are precisely assigned per user and team. This granular control significantly mitigates the risk of unauthorized access or accidental exposure of sensitive development resources.
H2: Streamlined Operational Controls for Uninterrupted AI Workflows
The gateway enables the application of essential operational controls, including data protection measures, usage limits, and policy checks, as AI interactions occur in real-time. A Universal API ensures that traffic is distributed efficiently across various AI providers, guaranteeing that quotas are strictly enforced and that potential outages do not stall critical AI-driven pipelines. This resilience is paramount for mission-critical AI applications.
H2: Unshakeable Agent Identity Security
Establishing trusted identities for AI agents is crucial for secure and auditable AI operations. Prisma AIRS AI Gateway achieves this by binding a verifiable, ephemeral identity to agents at the point of execution. The gateway then acts as an enforcement point, ensuring that only authenticated agents making approved calls can interact with AI models and services. This robust identity framework is essential for preventing unauthorized AI actions and maintaining accountability.
H2: Advanced Runtime Security for Proactive Threat Mitigation
Powered by Prisma AIRS AI Runtime Security, the gateway provides a critical layer of defense by inspecting every prompt and response inline. This dynamic analysis capability is designed to prevent the exfiltration of source code, secrets, and customer data from the network. Furthermore, it actively neutralizes sophisticated prompt injection attempts, aligning with the critical security guidelines outlined in both the OWASP LLM Top 10 and the OWASP Top 10 for Agentic Applications. This proactive threat mitigation is vital in safeguarding against the evolving landscape of AI-specific attacks.
The AI Control Plane for the Enterprise: Built for Scale and Resilience
Prisma AIRS AI Gateway is not merely a point solution; it is a foundational element of the enterprise AI infrastructure. Unlike individual tools that may focus on filtering text strings or routing single API calls, Prisma AIRS AI Gateway is engineered to handle the immense volume and complexity of real-world enterprise AI workloads.
"Most products simply fail at scale," commented [Executive Name and Title, e.g., Chad Kinzelberg, EVP of Palo Alto Networks’ Security Business Group]. "Prisma AIRS AI Gateway is built on an architecture that has been rigorously tested in the most demanding enterprise environments, supporting their ever-evolving AI workloads. We are providing the robust, scalable foundation that organizations need to confidently embrace the AI-driven future."
The platform is built upon an architecture proven in the most demanding enterprise settings, designed to evolve alongside the dynamic nature of AI technologies. This ensures that organizations are not investing in a solution that will quickly become obsolete.
Looking Ahead: Building the Agentic Enterprise Securely, Together
The agentic enterprise is no longer a future concept; it is a present reality. Developers have already embraced AI agents, and their integration into business processes is rapidly accelerating. As a company recently recognized by Gartner as the "company to beat" in AI Security Platforms, Palo Alto Networks is uniquely positioned to guide organizations through this transformative period.
Prisma AIRS AI Gateway represents a significant leap forward in enabling organizations to harness the power of AI responsibly and securely. By providing a unified control plane for governance, identity, and runtime security, Palo Alto Networks is empowering enterprises to accelerate their AI adoption without compromising on the critical requirements of security and control.
Organizations are invited to learn more about Prisma AIRS AI Gateway and register for an upcoming webinar to witness its capabilities in action. The journey towards an agentic enterprise is underway, and with Palo Alto Networks, that journey can be built securely, together.
