On June 2, 2026, a new Executive Order on AI was issued. Promoting Advanced Artificial Intelligence Innovation and Security signals a continued shift in how the federal government approaches AI innovation, security, and adoption. The order reinforces a familiar theme: The United States must move quickly to lead in AI, without compromising national security.
For federal agencies, this Executive Order is less about introducing entirely new concepts and more about accelerating action. It builds on prior guidance while placing sharper focus on operationalizing AI securely, strengthening cyber defenses, and deepening collaboration with industry.
Here’s what agencies need to know.
Why this Executive Order matters now
AI is no longer emerging technology in government, it is already embedded in mission systems, cybersecurity tools, and data analysis workflows. As adoption grows, so does risk.
The Executive Order acknowledges both sides of that equation. It highlights AI as a driver of economic growth and national strength, while also recognizing that advanced AI capabilities introduce new attack surfaces, new forms of exploitation, and new national security considerations.
For agencies, this creates a dual mandate. Move fast on AI adoption, but harden systems in parallel to ensure systems, data, and models are resilient against increasingly sophisticated threats.
Key actions agencies should expect
The Executive Order outlines several near-term actions, many within 30-60 days, that will directly impact federal agencies.
Immediate prioritization of cyber defense
Agencies, particularly those operating National Security Systems and civilian infrastructure, are directed to rapidly prioritize cyber defense. Within 30 days the order also calls on the Secretary of Homeland Security, through CISA and in consultation with OMB, to release Binding Operational Directives (BODs) aimed at strengthening protections across federal environments. Agencies should anticipate accelerated timelines for implementing cybersecurity controls, expanded use of AI enabled defensive tools and increased expectations for visibility across systems and data.
Expansion of AI-enabled cybersecurity capabilities
The order calls for the expansion of federal programs and services that leverage AI for defense. It also emphasizes improving access to advanced cybersecurity tools, not just for federal agencies, but also for state, local and critical infrastructure organizations.
Creation of an AI cybersecurity clearinghouse
One of the most notable provisions is the establishment of an AI cybersecurity clearinghouse. Within 30 days of the order, the Secretary of Treasury, in consolidation with the National Cyber Director, NSA and CISA must form an AI cybersecurity clearinghouse working with AI industry and critical infrastructure operators to:
- Coordinate and deconflict scanning for software vulnerabilities.
- Discover and validate such vulnerabilities.
- Coordinate and prioritize remediation and distribution of vulnerability patches.
In practice, this could significantly improve how vulnerabilities are identified and addressed, particularly in complex environments where software supply chains and AI models intersect.
Secure deployment of frontier AI models
The Executive Order introduces the concept of “covered frontier models”, advanced AI systems that meet certain thresholds of capability and potential risk. Within 60 days, Treasury, NSA, CISA and others must develop a classified benchmarking process to assess the advanced cyber capabilities of AI models and determine the threshold at which an AI model should be designated a covered frontier model. They must also design a voluntary framework that would allow AI developers to engage with the federal government before releasing advanced models. Through the framework, developers can provide up to 30 days of pre-release access to covered frontier models for evaluation and security testing.
The order stops short of imposing mandatory licensing or pre-approval requirements, reinforcing a commitment to innovation while encouraging responsible deployment.
Stronger enforcement against AI-enabled threats
The order prioritizes enforcement against criminal activity involving AI, including unauthorized system access, data breaches, and fraud. This underscores the reality that AI is not just a defensive tool, it is also being used offensively by threat actors. Agencies must be prepared to defend against AI driven attacks that are faster, more automated, and harder to detect.
Investments in the federal workforce
Recognizing that technology alone is not enough, the order also calls for expanding federal cybersecurity hiring pathways, including strengthening programs to recruit and place cybersecurity professionals across government.
Why this matters for federal agencies
The Executive Order reinforces a few key realities for federal leaders:
- AI adoption and cybersecurity are now inseparable. Agencies can’t advance one without the other.
- Speed matters. Many of the directives carry aggressive timelines, requiring rapid execution
- Collaboration is essential. The government will increasingly rely on partnerships with industry to stay ahead of threats.
- Visibility and control are foundational. Agencies must understand where AI is being used, how data flows through systems, and where risk exists.
How Netskope can help
Netskope helps federal agencies translate policy into action by providing a unified, data-centric approach to security across cloud, SaaS, web, private apps, and AI.
With Netskope, agencies can gain full visibility into AI usage across users, applications, and emerging AI services; protect sensitive data in real time with guardrails to prevent exposure through AI prompts, outputs and model interactions; detect and block AI specific threats such as prompt injection and jailbreaks; and enforce zero trust principals consistently to ensure least privilege across all environments. With red teaming capabilities, agencies can continuously validate AI systems against real-world attack techniques, reducing operational and mission risk before vulnerabilities ever reach production.
For more information about how Netskope can secure AI for federal agencies, read more here.
