Both can be true at the same time. Big AI companies can be trying for regulatory capture while there may be real dangers, both short-term as well as long term, perhaps even existential dangers.
Why do people seem to think evidence for one of these is counter evidence for the other?