The old fear of deepfakes was that "people wont be able to tell real from fake." That initial fear seems to have subsided and the new fear is "AI will allow malicious actors to manufacture so much believable content that truth as a concept will evaporate."
I think such a situation is impossible and almost even nonsensical. Reality itself is full of human agents lying, obscuring and misleading about their own internal processes. How could an organism like ourselves be "misled" by artifacts produced by devious beings?
It just seems to imagine that humans can be short-circuited by some malicious input and never adapt or change. It seems to me we have been bombarded by malicious inputs for 10000 generations, why are these new inputs so dangerous?
