x
A Defense of Work on Mathematical AI Safety — AI Alignment Forum