I love Schneier and his legacy of activism, and I read _Applied Cryptography_ on my own in high school, with great relish. Yet this article specifically is a t

I love Schneier and his legacy of activism, and I read _Applied Cryptography_ on my own in high school, with great relish. Yet this article specifically is a tone deaf piece by someone who still subconsciously believes we are in the late nineties, with a contiguous and coherent "public interest," and culture wars as sideshow rather than main event. As it actually stands, any nationalized AI product would be the ultimate battlespace. What it could and could not say would change with the administration. Imagine if your chatbot therapist, powered by nationalized AI, suddenly became an abortion snitch, for example. What Schneier actually does do here, albeit unsuccessfully, is surface the actual problem, which is that we might not agree with each other enough to make AI economically feasible, if indeed OpenAI and Anthropic do not work as for-profit corporations.