zlacker

Sam does believe in safety. He also knows that there is a first-mover advantage when it comes to setting societal expectations and that you can’t build safe AI by not building AI.