Pillsbury
Lawyers
Capabilities
Insights
Alert
By Shruti Bhutani Arora, Anne M. Voigts, Brian E. Finch, Mark L. Krotoski, Ya’ara Z. Barnoon, Nathan D. Banks
Alert
By Shruti Bhutani Arora, Anne M. Voigts, Brian E. Finch, Mark L. Krotoski, Ya’ara Z. Barnoon, Nathan D. Banks
06.10.26
On June 2, 2026, the White House issued an executive order (EO) titled Promoting Advanced Artificial Intelligence Innovation and Security, establishing a federal policy framework that prioritizes AI-driven cybersecurity while preserving an innovation-first, voluntary regulatory model. The EO was reportedly edited to respond to industry concerns and is framed as a pro-innovation measure. The EO expressly states that it does not create a mandatory licensing, preclearance or permitting requirement for the development, release or distribution of AI models. At the same time, it strengthens cybersecurity across federal and critical infrastructure systems using AI-enabled tools, seeks to protect intellectual property and technology from adversarial exploitation, and directs closer coordination with the private sector on AI security risks and mitigation strategies. Notably, the federal government is increasingly treating advanced AI capabilities as a cybersecurity and national security issue, particularly where frontier models may be capable of discovering, validating or exploiting software vulnerabilities at scale.
The EO also follows a growing trend at the state level, where states such as California and New York have begun imposing frontier AI-specific governance, transparency and incident-reporting obligations on frontier developers, including requirements aimed at assessing and mitigating risks of catastrophic or “critical harm.”
Key provisions of the EO include:
Why This Matters for AI Governance
Although the EO stops short of creating a mandatory approval regime for frontier AI models, it makes clear that advanced AI capabilities are increasingly being viewed through the lens of cybersecurity and national security.
For developers of frontier AI models, the order puts particular emphasis on model capability, secure pre-release engagement with the federal government and governance over early access to covered frontier models. The EO directs federal agencies to develop a classified benchmarking process to assess advanced cyber capabilities and to design a voluntary framework through which developers may engage with the government about whether models under development qualify as covered frontier models. It also contemplates federal access to covered frontier models for up to 30 days before release to other trusted partners, subject to confidentiality, cybersecurity, insider-risk and IP protections.
The EO is also relevant to companies deploying AI-enabled cybersecurity tools or AI agents that interact with critical infrastructure. The EO directs federal agencies to expand access to AI-enabled cybersecurity tools for government agencies, state and local authorities, and critical infrastructure operators, and to establish an AI cybersecurity clearinghouse focused on vulnerability scanning, validation, remediation and patch distribution. It also directs the Attorney General to prioritize enforcement against actors who use AI, including AI agents, to unlawfully access computers, data or information.
Key Compliance Considerations
Companies developing or deploying covered frontier models, AI-enabled cybersecurity tools or AI agents that interact with sensitive systems should consider the following steps:
Board-Level Questions
Boards should consider directing management to evaluate several key questions:
Pillsbury helps clients navigate AI governance, data privacy, cybersecurity and regulatory compliance, including model risk assessments, cybersecurity and regulatory litigation, data-use governance, incident response, and engagement with emerging federal and state AI requirements.
Alert 06.09.26
Alert 12.19.25
The Executive Order outlines a federal strategy to limit, challenge and potentially preempt state AI laws, intensifying the debate over who sets the rules for AI in the United States.
© 2001–2026 Pillsbury Winthrop Shaw Pittman LLP
