Why the GSA OneGov Agreement Is a Game-Changer for Federal Cybersecurity

The mission to modernize government IT is accelerating at lightning speed, largely thanks to the transformative power of artificial intelligence (AI). Federal agencies are strategically leveraging AI to boost efficiency, enhance citizen services, and strengthen national security – a vision fully supported by the administration’s AI Action Plan.

At Palo Alto Networks, we are all-in on helping agencies deploy AI bravely and securely. Because the challenge isn't just about using AI for cyberdefense, but also about defending AI itself. We appreciate the U.S. General Services Administration (GSA) recognizing the critical need for scalable, efficient solutions.

That is precisely why the GSA OneGov Initiative is a massive, game-changing step forward. We are proud to be the first pure-play cybersecurity vendor to secure a OneGov agreement with the GSA. This strategic alliance simplifies and standardizes the process for agencies to access our world-class, AI-powered security platform, ensuring security is foundational to this crucial modernization mission.

The Wake-Up Call: The Silent Threat of AI Agent Corruption

If you needed a clear sign that AI has fundamentally shifted the cybersecurity landscape, our own Unit 42 research provides it. The new reality isn't just about hackers using AI in their attacks; it’s also about how internal AI provides another attack surface for threat actors.

The most insidious new threat we've observed is AI Agent Smuggling, where malicious attackers use AI agents to exploit other agents. Our Unit 42 research highlights two major vectors:

• Indirect Prompt Injection: A security risk in LLMs where a user crafts input containing deceptive instructions to manipulate the model’s behavior, which can lead to unauthorized data access or unintended actions.
• Agent Session Smuggling: Exploit vulnerabilities in agent-to-agent communication, injecting malicious instructions into a conversation, hiding them among otherwise benign client requests and server responses.

This confirms our core belief as stated in a recent secure AI by Design blog: The AI ecosystem (the models, data and infrastructure) is now a complex, expanding attack surface that traditional perimeter defenses were simply not designed to protect.

As I’ve said before, “If you’re deploying AI, you must deploy AI security.”

Secure AI by Design: A Strategic Alliance with GSA

The GSA’s OneGov Initiative aims to streamline procurement and drive down costs by leveraging the purchasing power of the entire federal government. This is more than an agreement; it’s a direct response to the call for a "secure-by-design" approach to federal AI adoption. This agreement simplifies and standardizes the process for agencies to access our world-class, AI-powered security platform, ensuring that security is foundational, not an afterthought. It provides industry leading AI security tools into the hands of our cyber defenders today.

Under the Hood: Technical Capabilities for the AI Ecosystem

To counter the autonomous threats we’re seeing, we provide a platform that protects the entire AI lifecycle, from the developer's keyboard to the data center.

1. Runtime Protection for AI Workloads

Securing the AI supply chain requires visibility across every stage, especially during runtime when models are processing sensitive data.

• Prisma^® AIRS delivers comprehensive security for the entire AI lifecycle, in one unified platform. It allows organizations to deploy traditional apps as well as AI applications, models and agents with confidence by reducing risk from misuse, data loss and sophisticated AI-driven threats. Prisma AIRS provides a clear, connected view of assets in multicloud environments, so teams can eliminate silos, accelerate responses, as well as scale cloud and AI apps securely.
• Our Cloud-Native Application Protection Platform (CNAPP) has achieved the FedRAMP High designation, making it the preferred Code to Cloud solution to secure the entire application lifecycle from development to runtime. Our industry-leading CNAPP eliminates silos to deliver comprehensive visibility and best-in-class protection across multicloud environments.

2. Protecting Users and Data at the Edge

Even the most advanced AI defenses are undermined if users accessing applications and data are left vulnerable outside corporate security boundaries. The explosive growth of generative AI tools and the unseen behavior of AI agents are amplifying data exposure risks.

• Prisma SASE (secure access service edge) secures all users, apps, devices and data, no matter where they are and no matter where applications reside.
  • Prisma Access (FedRAMP High Authorized) and Prisma Browser (FedRAMP-Moderate Authorized) integrate security capabilities, like zero trust network access (ZTNA), secure web gateway (SWG) and cloud access security broker (CASB), to provide a unified policy framework and a consistent user experience.
• This approach helps agencies outpace the speed of AI-driven threats, safeguarding critical data and simplifying operations for a frictionless user experience. It ensures that the human element interacting with the AI is protected by the most stringent security controls available.

Deploy AI Bravely

The GSA OneGov agreement is a pivotal moment that provides federal agencies with the cost-effective, streamlined access they need to deploy AI with confidence. By leveraging our unified, AI-powered platform, government organizations can stop reacting to threats and start building secure-by-design AI environments. We are committed to remaining a key partner in this strategic initiative and helping the government achieve its mission outcomes safely.

For more information and access to promotional offers for new contracts signed on or before January 31, 2028, federal agencies can visit the GSA OneGov website.

The post Securing the AI Frontier appeared first on Palo Alto Networks Blog.

Modernizing Vulnerability Sharing for a New Class of Threats

In cybersecurity, vulnerability information sharing frameworks have long assumed that conventional threats exploit flaws in software or systems, and they can be resolved with patches or configuration updates. AI and machine learning (ML) models upend that premise as adversarial attacks, like poisoning and evasion, target the unique way AI models process information. Consequently, the risks for AI systems include tactics like model poisoning (from evasion attacks) in datasets and training, which are not conventional software vulnerabilities. These new vulnerabilities fall outside the scope of traditional cybersecurity taxonomies like the Common Vulnerabilities and Exposures (CVE) Program.

There is a need to bridge the gap between the existing cybersecurity vulnerability sharing structure and burgeoning efforts to catalog security risks to AI systems. Provisions in the White House AI Action Plan, which Palo Alto Networks supports, call for the creation of an AI Information Sharing and Analysis Center (AI-ISAC), reinforcing the importance of addressing that disconnect. This integration is essential, as leveraging the existing, widely adopted cybersecurity infrastructure will be the fastest path to ensuring these new standards are accepted and operationalized.

Established Construct for Vulnerability Management and Disclosure

The global cybersecurity community relies on a mature infrastructure for sharing standardized vulnerability intelligence. Central to this ecosystem is the CVE List, established in 1999 as the authoritative catalog of cybersecurity vulnerabilities. Through CVE IDs and a network of CVE Numbering Authorities (CNAs), this framework enables consistent vulnerability documentation and disclosure.

Similarly, the Common Vulnerability Scoring System (CVSS) provides standardized severity assessments, allowing security teams to prioritize responses. Together with resources like the National Vulnerability Database (NVD) and CISA’s KEV Catalog catalog, these tools form the backbone of global vulnerability management, information sharing and coordinated disclosure.

Why AI Breaks the Traditional Model

While this infrastructure has served the cybersecurity community effectively for over two decades, it was designed around traditional threat models that AI systems substantially upend. Attacks on AI systems represent a critical departure from traditional cybersecurity threats as they operate insidiously, subtly corrupting core reasoning processes, causing persistent, systemic failures, some of which only become evident over time. Most traditional cybersecurity tools are not equipped to recognize those breakdowns because they assume deterministic behavior and rules-based logic. AI systems defy those assumptions because AI is probabilistic, not deterministic. Consequently, attacks on AI models may remain hidden for extended periods.

Unlike traditional cybersecurity threats that target code, adversarial AI attacks target the underlying data and algorithms that govern how AI systems learn, reason and make decisions. Consider the following predominant adversarial attack methodologies on machine learning:

• Poisoning attacks inject malicious data into training datasets, corrupting the model's learning process and creating deliberate vulnerabilities or degraded performance.
• Inference-related attacks exploit model outputs to extract sensitive information or learn about its training data. This includes model inversion, which reconstructs sensitive data from the model's outputs, as well as membership inference, which identifies whether specific data points were used in training.

The expansion of existing security frameworks and programs is necessary to cover the enumeration, disclosure and downstream management of security risks to AI systems.

Advancing AI Security Through the AI Action Plan

In July, the Administration unveiled the AI Action Plan, an innovation-first framework balancing AI advancement with security imperatives. The Plan prioritizes Secure-by-Design AI technologies and applications, strengthened critical infrastructure cybersecurity and protection of commercial and government AI innovations.

Notably, it recommends establishing an AI Information Sharing and Analysis Center (AI-ISAC) to facilitate threat intelligence sharing across U.S. critical infrastructure sectors and encourages sharing known AI vulnerabilities, “tak[ing] advantage of existing cyber vulnerability sharing mechanisms.” These provisions affirm that AI security underpins American leadership in the field and, where possible, should be built upon existing frameworks.

Redefining Boundaries for AI Threats

To position the CVE Program for the AI-driven future, Palo Alto Networks is engaging directly with industry and program stakeholders to chart the path forward. Traditionally, the CVE Program serves as an ecosystem-wide central warning system. It provides a unified source of truths for security risks. A security risk catalog and identification system are needed for AI systems, as they currently fall outside the traditional scope of the CVE Program that has focused exclusively on vulnerabilities rather than on malicious components. The historical aperture of the current CVE Program excludes harmful artifacts, such as backdoored AI models or poisoned datasets, which represent fundamentally different attack vectors, in turn creating security blind spots.

Securing AI’s Promise

The United States leads in AI innovation and must equally lead in securing it. As momentum builds behind the AI Action Plan and the establishment of the AI-ISAC, we have a critical window to shape information sharing frameworks of the future. The goal is to ensure that cybersecurity and AI security infrastructure advance in unison with the technology itself. Integrating new AI vulnerability standards into trusted frameworks like the CVE Program aligns with industry focus and needs. Through proactive, coordinated action, we can unlock AI’s full promise while safeguarding the models that are embedded in the critical systems on which our nation depends.

The post Bridging Cybersecurity and AI appeared first on Palo Alto Networks Blog.