MY TAKE: The Pentagon punished Anthropic for red lines it accepted from OpenAI hours later

It's about who controls the behavioral boundaries of an AI model that operates inside classified military systems. Claude was already deployed on ...