Gordon Seidoh Worley comments on Thoughts on “AI safety via debate”