Linda Linsefors comments on Towards more cooperative AI safety strategies