Karl von Wendt comments on The benefits and risks of optimism (about AI safety)