OpenAI Releases 'National Security Principles,' Clarifying AI Usage Boundaries in Government and Defense Sectors

CoinFeed July 9 news, OpenAI releases 'National Security Principles,' clarifying boundaries and usage norms for cooperation with governments in national security and law enforcement. The principles will apply to existing and future cooperation with agencies such as the U.S. Department of Defense, reiterating the prohibition of using its technology for mass domestic surveillance, commanding autonomous weapon systems, and high-risk automated decision-making. OpenAI states that it has established 'trusted access' cooperation with the U.S. and allies in cybersecurity and biosecurity, including cooperation with multinational cybersecurity agencies through the Daybreak program, and opening the biosafety model GPT‑Rosalind to some government agencies for public health and biodefense missions. OpenAI expresses support for legislation to set external constraints on high-risk military AI uses.