Twitter thread on politics of AI safety — AI Alignment Forum