Trump has signed an executive order to enhance U.S. cybersecurity using AI, creating a voluntary evaluation framework for advanced models.
Trump Issues Executive Order to Enhance AI-Driven Cybersecurity
President Donald Trump has signed a new executive order aimed at strengthening U.S. cybersecurity through the use of advanced artificial intelligence technologies. This initiative is designed to foster collaboration between federal agencies and major AI companies while establishing a voluntary framework for evaluating powerful AI models before their deployment.
The order, named "Promoting Advanced Artificial Intelligence Innovation and Security," directs federal agencies to accelerate the adoption of AI-based cybersecurity solutions and to set up an AI cybersecurity clearinghouse. Additionally, it will implement a system to identify advanced AI models that may affect national security.
In the executive order, Trump stated, "Advanced AI capabilities make our nation stronger, but also introduce new national security considerations that require coordinated action across executive departments and agencies." He emphasized the need for close collaboration with the industry to ensure rapid deployment of secure technologies against threats.
A classified review process will be established, led by the National Security Agency, to assess whether advanced AI systems qualify as covered frontier models. Developers can voluntarily submit these models for evaluation up to 30 days before they are released to trusted partners.
This executive order follows a delay in issuing a similar directive, as Trump previously expressed concerns that some aspects of the proposal could undermine U.S. leadership in AI in relation to China. Critics, however, argue that the voluntary nature of the compliance framework may not provide sufficient oversight for the powerful AI models it aims to regulate. J.B. Branch from Public Citizen called for more comprehensive federal legislation to ensure accountability and protection for consumers and workers.
The urgency for an AI-focused executive order has been heightened by concerns over Anthropic's Claude Mythos model, which has demonstrated the ability to detect software vulnerabilities, raising alarms about its potential national security implications. Despite these concerns, Anthropic announced plans to increase access to Claude Mythos through Project Glasswing, allowing tech and security firms, as well as government entities, to identify vulnerabilities before its public release, expected in the coming weeks.
The executive order also seeks to assure AI developers that it will not enforce a formal approval process for launching new AI models. It coincides with Trump's broader efforts to create a federal regulatory framework for AI amidst a wave of state-level legislation. Furthermore, the order includes provisions for stricter enforcement against the criminal use of AI, targeting unauthorized access to IT systems and illegal data acquisition. Recent prosecutions under the new Take It Down Act highlight the growing focus on AI-related crimes.
