New Trump executive order requests AI companies 'voluntarily' allow the White House to test the "advanced cyber capabilities of AI models” (www.techradar.com)

President Donald Trump has signed a new executive order requiring leading AI companies to voluntarily submit their models for cybersecurity testing to the White House prior to market release. This shift marks a notable change in the administration's approach to technology regulation, prompted by the unveiling of Anthropic’s Mythos Preview, an AI tool capable of uncovering significant software vulnerabilities. With major AI firms like Google and OpenAI expressing support for this initiative, the order aims to ensure the safe deployment of advanced AI systems in an ever-evolving cybersecurity landscape. The executive order directs various government agencies, including Treasury, Defense, and Homeland Security, to establish agreements for testing AI models. This initiative reflects a growing recognition of frontier AI as a strategic asset, comparable to advanced cyber tools, highlighting the importance of balancing innovation with safety. The convergence of regulatory philosophies from different administrations underscores a broader consensus on the critical role of AI in national security. As the industry engages in voluntary cooperation, the ongoing debate centers around whether security should be achieved through regulation or competitive advantage, shaping AI policy for the foreseeable future.