quetzal_rainbow comments on Towards more cooperative AI safety strategies