Caitlin Kalinowski, head of robotics at OpenAI, has resigned over concerns regarding a Pentagon contract involving AI deployment on classified systems. Kalinowski reportedly warned that the proposal lacked adequate safeguards against domestic surveillance and lethal autonomous weapons. OpenAI maintains it has strict boundaries on military applications, but the resignation underscores tensions between Silicon Valley's AI ambitions and national security priorities. In a related development, the Pentagon has designated Anthropic as a "supply-chain risk," restricting defense contractors from using certain versions of its Claude AI model. This rare designation for a domestic tech firm highlights growing disagreements over AI safety and military use, as defense agencies increasingly explore advanced AI systems for sensitive operations.