Larks comments on Where Iā€™m at with AI risk: convinced of danger but not (yet) of doom