White House directs faster AI adoption across US national security agencies

The White House has issued National Security Presidential Memorandum 11, setting out a new policy for AI use across the US national security enterprise.

The memorandum, published on 5 June 2026, directs defence, intelligence, homeland security, energy, and other national security agencies to accelerate AI adoption while maintaining oversight linked to civil liberties, privacy, reliability, and accountability.

The policy is built around four pillars: adoption, adaptation, assurance, and accountability. Agencies are instructed to remove barriers to AI deployment, adapt commercial and open-source AI systems where appropriate, and ensure that adopted systems are reliable, robust, steerable, and controllable.

The memorandum also states that AI used by national security agencies must not be developed or used to censor speech, embed ideological bias, or conduct unlawful surveillance. Commanders and agency heads remain responsible for ensuring compliance with legal and constitutional requirements.

Several deadlines are included. Within 90 days, the Department of War must update its directive on autonomy in weapons systems, and national security agencies must receive updated governance guidance for AI use in national security systems.

Within 120 days, agencies must update procurement processes to enable faster onboarding of advanced AI models from multiple vendors. Officials are also tasked with developing a roadmap for secure access to advanced computing resources, including high-security AI computing facilities and an AI test range for national security use cases.

The memorandum also calls for partnerships with private companies to protect advanced AI technologies from threats such as malicious model distillation, as well as joint AI data and model exchanges across national security agencies.