There will absolutely be questions and discussions of model welfare within companies ie new company policies for how to treat (or not treat) AI and models. Anthropic’s terms of use change was just an early mover.