You have no more basis for claiming this than AI reseachers have for claiming the opposite. More research is required, and it will be a well spent effort if the

You have no more basis for claiming this than AI reseachers have for claiming the opposite. More research is required, and it will be a well spent effort if the probability of a bad outcome in AI development is nonzero.