“Leopold Aschenbrenner, formerly of OpenAI, has argued that as AI becomes critical to national security, governments will not allow private companies to control it and will seek a more integrated, nationalized approach.”