Trump's Directive: America Must Act Swiftly Before It Is Too Late

The United States is set to accelerate the integration of AI to strengthen national security protections. This initiative is part of a directive from President Donald Trump, intended to ensure the nation does not fall behind its global competitors.

The Trump administration will request leading AI developers to voluntarily submit their models for government cybersecurity testing before they are released to the public. “Under my administration, the US can and will responsibly accelerate the use of AI across all intelligence and warfare domains in accordance with American values,” Trump stated in a national security memorandum, as reported by Reuters.

Trump has also granted Defence Secretary Pete Hegseth a 90-day window to update directives regarding the autonomy of weapon systems. These guidelines aim to ensure that the adoption of AI systems respects the established chain of command. Furthermore, the directive ensures that AI will not be developed or utilised by national security agencies to censor freedom of speech, nor for unauthorised or unlawful surveillance.

Michael Kratsios, Director of the White House Office of Science and Technology Policy, explained that the memorandum will accelerate AI adoption across various vendors, prevent systemic failures, and update Department of Defence guidelines. He emphasised the need to ensure that no entity can disable or degrade the quality of AI systems relied upon by soldiers without prior approval.

This directive follows recent friction between the AI company Anthropic and the Pentagon. The US government has identified Anthropic as an official supply chain risk due to the company’s refusal to withdraw its ban on using the Claude AI model for autonomous weapons and US mass surveillance. According to the Pentagon, authorities may utilise technology as required, provided it remains in compliance with national law.