A good speculation. Among the conclusions is: "In a certain sense, LLMs do not represent a groundbreaking change in terms of the capabilities of a well-funded

A good speculation. Among the conclusions is: "In a certain sense, LLMs do not represent a groundbreaking change in terms of the capabilities of a well-funded intelligence agency. They merely lower the cost." In a sense, this recapitulates the productivity pitch. They also point out the influence operations, although I am not sure whether such things will make any difference as the information space is already so crowded with garbage.